1. Dirty Secrets of Clean Sound with Metacast Hosts Ilya Bezdelev and Arnab Deka

It takes so much time, it feels like a second job to produce one episode in a month. So I did a time-shape. So you decided to add on to that work and collect metrics about it. You should at all costs avoid air pods. I love my air pods, right? But I guess I'll just keep using them to listen to podcasts while walking my dog.

Hi everyone, thanks for tuning in to our podcast. That's the very first episode of Metacast. It's a podcast for podcasters about all things podcasting. That's why we call it the Metacast. We think this podcast will be interesting to aspiring podcasters and also current podcasters and everybody who is a fan of podcasts like my friend Arnab, who is a co-host with me today.

Hello, my name is Ilya Bezdelev. I'm a fan of all things audio, podcasting, audio production, audio books, music, you name it. Also, to this podcast I work as a product manager at Google Maps. I'm also a hobby musician and I host another podcast about people and technology. In this episode we are going to talk about my process of creating podcasts, which is also pretty meta. And the co-host of Metacast is Arnab, my longtime friend and a former colleague.

Hello everybody. So like Ilya said, my primary mode up consuming information is also audio, audible, lot of podcasts. I got started with cereal, I don't know, four or five years ago. And now every chance I get like while walking a dog or doing dishes, I'm listening to a podcast. I am a software engineer by profession until this summer I was working with AWS. And then recently I left my job and I'm experimenting with a few things.

One of those things is around podcasting. So I thought, hey, why don't I jump right into the podcasting space by trying to co-host a podcast. Yeah, so that's Arnab's very first experience recording a podcast. So let's give him a round of applause. Thank you, Ilya. By the way, we'll get into a lot of the tips and tricks later today. But I wanted to quickly mention one interesting thing.

Do you ask me to sit in the closet while recording the first episode since we don't have a recording studio or anything? So here I am in my bedroom at 7 a.m. in the morning recording the very first episode. And I hope my dog doesn't become very inquisitive and try to figure out what's happening in here.

You know, when we got our house, I'm like, I'm going to have a studio so that I can record podcasts and music with high quality. But it turned out that it's actually very expensive to convert a room into a studio.

So guess where I am sitting right now in my closet, the bedroom. And I sent my wife and my kiddos somewhere else so that they don't come and investigate and also had to shut down the air conditioning unit because otherwise there is a lot of noise. And we're going to talk about that as well. Hopefully these episodes will serve two purposes. First, you will learn about me and Arnab a little bit and it will inspire you to listen our fish episodes.

And second, you will hopefully get something useful out of this show because I started my other podcast about a year ago. And you know, I hit quite a few challenges and it wasn't easy. So I can totally empathize with people who are sitting now listening to this and maybe scared to start the podcast or don't know where to begin or think they don't have the skills or don't like their voice. We are going to talk about all of that today. Let's just dive right in.

Yeah, and this is more or less the format that we'll be following for our future episodes to where we'll have a guest, maybe your favorite podcaster and we'll learn about their story and the tips and tricks and stuff they use. Yeah, all right. Let's get started. So Ilya, tell me about your podcast. What's it about? Who listens to it? What do you talk about? Is it like narrative or interview style?

The podcast I host is called IT Samurai. Actually, it's called IT Samurai because it's in Russian and that's about people in technology. I do in Russian because I did want to move back to Russia.

Now it's kind of off the table because of the current events is October 2022. We won't go into all of that, but it's kind of off the table right now. Initially when I started that, I really wanted to build my personal brand, build some following, build the reputation. So when I go back to Russia, I already have this kind of base on which I can build either a business or a career or maybe even make money with the podcast itself.

It was also great for networking opportunities and also just sheer fun of talking to people who work in the same industry as I am. So I had designers, I had engineers, I had engineering managers on my podcast. Actually, one kind of profession I never had as a guest is a product manager.

I guess I want to keep the monopoly on the product manager profession. Actually, I just like I had one product manager on my show. People who listen to that, mostly people who are obviously speaking Russian language. The majority of my listens comes from Russia and neighboring countries. Some listens from outside of the former US subblock. So you're saying that most of the listeners are physically in Russia. How about all the Russian immigrants in the US and everywhere else?

According to statistics, most of the listeners come from the Russian IPs. So yes, they are in Russia, some in Ukraine, some in Belarus, some in Kazakhstan. Maybe about 10% comes from the rest of the world. US, some people in Europe. I think what happens here, I'm extrapolating my own experience. When I moved out of Russia 15 years ago, I started consuming most of my content in English.

The only podcast I listened to in Russian are some kind of news podcasts and also entertainment, where language actually matters. But when it comes to any professional podcast, knowledge and information, I prefer to listen in English and I suppose the same thing happens to a lot of immigrants, especially in our world, in the information technology, because we have to speak English. Yes, so people just switch to another language as they immigrate.

That makes sense. In technology, it's predominantly English, yeah. Did you ever think, should I started in Russian or English, or did you ever figure out that part? Starting podcast was a challenge of its own, because I never really liked my voice, the way it sounds. So that was the very first barrier I actually had to overcome to even start doing something. I didn't really question Russian versus English. I knew English would give me much broader reach, much broader audience.

But because I wanted to move to Russia, I'm like, okay, so let's start it in Russian. It will be easier for me because it's my native language. And also people in Russia are more hungry for the content I can give them compared to people in the US, for example, where there are lots of people like me. But there aren't many people like me, speak in Russian. So that's why it was a better niche for me.

I also had a blog in Russian, so that gave me the platform for distribution of the podcast, which also played the role here. Makes sense. So for people like me who's interested in listening to your podcast, but I don't speak or understand any Russian, have you heard of people interested from people like me? What do you do for them? Yeah, the fun part here is usually when I introduce myself to a new colleague, for example, at work.

And I mentioned something about the podcast because they see all my gear in the camera. And they would be asking about what podcast is about, if they can listen to it, if you can send a link and I tell them it's in Russian. And they would go like, oh shoot, I really wanted to listen to it. And I'm always curious if they say this out of politics. Or if they really want to listen to it. I know you would. Yeah, we know each other for quite a long time, but other people I don't know.

Well, the thing is they can't. I suppose they can use some audio translation software, but that would probably sound terrible. And most of the nuances, language specific jokes that will not translate. And one of the things about my podcast is that when I say it's about people and technology, that's because it's about people. There are lots of podcasts about technology. You can read about technology anywhere. What I really want to get out of my guests is their human side.

It's to understand how they think. And not just in terms of mental models or this kind of structured stuff. More like what they struggle with. How do they overcome this? So what are some of the emotional things that they experience? Just to give you an example, I had one guest who was open about her bipolar disorder. And we talked about that. And if they were gonna therapy the chooses. And then somebody else wrote to me.

And he's like, listen to your interview with this person. And I have a bipolar disorder too. And this is just so helpful. This is the side of people I want to show there, the human side. And that's why if you actually translate it into another language, a lot of that will be lost. That's a really solid motivation that you get, right? For keeping on doing it in your native language. Because you won't get that sort of emotions if it's not your first language.

I've not heard any podcast that is not in English. How different is it to do a podcast in a non-English language? I'm assuming there are some challenges. The main challenge I have experienced is the RISIS tool called Descript that allows you to import your audio file, your recording file. And it would transcribe the audio into text. And then you can just edit text like a document. And it will cut out the audio or rearrange the audio.

And then you'll also cut out arms and ooms and other kind of things. And guess what? It only works in English. Without that, I have to do all of that manually. I have to do more work myself to produce my podcast. Also, for English-picking podcasts, there are a lot more opportunities to outsource the audio production, where you can hire somebody and they will do that work for you. In Russian, the pool of those sound engineers is much smaller.

And it's not a mechanical job, so you do have to understand what's being said and understand the emotions and stuff behind it to do a good job about it. So make sense that you need a native speaker to do this job. Exactly. Because one of the things in the podcast that's very, very important is the pace of speech. If you shorten pauses too much, the speech starts to sound unnatural. It's more efficient, but it's not very effective.

Like you said, give through a starter. How do you know it's starter if you don't speak a language? You can do it. I think the only thing you can do mechanically is to make the audio quality sound good. But now, with only plugins and filters, if you spend some time learning those and some money buying those, actually you don't necessarily need a sound engineer. You can do it all yourself with those plugins. It's fairly straightforward.

Another thing I would say about other languages, and I don't know if that's a bug in the software or it's specific to Russian speech patterns, I'm using a software called IZOTOPE RX9 with all the different effects. One of the effects is it removes breaths. Microphones pick up a lot of breathing, and it's pretty terrible. So I use the software, so I don't have to remove those manually. And for some people, I think two or three people I had on my show, it just wouldn't remove any breath.

So somehow it is unable to pick up breaths from the rest of the speech, and so that also introduced more work for me. So yeah, I guess language introduces more technical sound. Yeah, when you asked me to get this mic, right, because we were going to start this podcast, the Blue Yeti thing, and I put it on and tried to like play with the knobs, and all I was surprised to hear the amount of background noise that it picks up.

If there is a washing machine going on in the basement, which is two floors below here, it still picks up the background noise from there. It's crazy, yeah. Dude, if you have a fluorescent light, it pick up the light noise. Nice. Nice. Yeah. And I think what you said about the state of software and non-English languages. By the way, Ilya and I were working in AWS about five years back, I guess, together, maybe even more. So we worked together for five years, and it started seven years ago.

At that time, I think if you remember, we were working on the chatbot, and we were using NLP software, and it was the same. We found the same things, right? For English, it does a pretty decent job. If you put in anything else, it's like, yeah, I can't understand. Exactly. I think that may have improved a bit over the last five to seven years, but that differential between English and non-English probably still exists quite a lot.

I guess the language is like Spanish, which I need, where I have a very large native speaker base. They will probably improve much faster, because there is a bigger market there, so that follows the money. Whereas with some languages with smaller populations, it will get there. It will just take time. You mentioned your motivation to start the podcast, but how did you actually decide to start the podcast? Like, what triggered that?

I wanted to do a podcast ever since I listened to my first podcast, maybe ten years ago. But I always had this thing, what am I going to say? And I also really didn't like my voice. I don't know if you remember those old video camcorders from the 1980s, where you put like a VCR tape, it records you. We had one of those when I was a kid, and I hated listening to me speaking. Like, I would try not to speak when somebody was filming. So I had this thing from my very early days. I hated my voice.

So I would have started my podcast sooner if I didn't have that. And then I think what finally changed the tide for me is somebody who I know from Russia. He invited me as a guest on his podcast. It's podcast about Russians living abroad. And I'm like, I don't know. And then we recorded it. I listened to the result. And I'm like, okay, actually, my voice sounds fine. It's just fine. Well, the trick there was that the edit is quite heavily.

But listening to that gave me a lot of confidence that even if I don't like my voice, even if I don't like my pattern of speech, that actually can be adjusted with post production. And also it's not as terrible. That was the trigger for me to actually start doing that. I mean, I hesitated for a while, maybe like six months or so. And then I remember it was last spring, spring 2021, you were in Hawaii. And I'm like, I got to start doing it. I drew artwork on my iPad with an Apple pencil.

I was getting it back then. And I bought an expensive mic. What was it? It was a sure MV seven. If you watch, let's say Joe Rogan. And you see the microphone he has and many other people have the same kind of microphone. That's a sure SM7B. It's like a canonical microphone for radio broadcasting. And for podcasting, it costs 400 bucks. And also you need to buy an audio interface, which is another couple hundred bucks. That's pretty expensive.

And then sure was following the money. I mean, it was following the trend of podcasting. And it introduced a cheaper version of SM7B called Sure MV seven. That has a USB, it plugs directly into your USB port in your computer. And it was $250, which is still pretty expensive for a mic. If you don't know if your podcast is going to work, but I bought a mic. I'm like, I'm going to commit to it. So by spending 250 bucks. So that mic arrived to Aho. And then I tried to record an narrative episode.

I had some experience with audio recording because of my music experiments. So I knew the audio recording site. I wouldn't say super well, but I knew it well enough. Yeah, I put together like a plan of what I want to talk about. And I rumbled for two hours. I never listened back to what I recorded because I thought it was terrible. There was a lot of stutter and stumbling and re-recording. And I thought I was speaking to quietly. I didn't like it at all.

Actually, I read an advice. I think it was NPR's podcast startup guidebook where they said when you first start record two or three episodes and just throw them out. Because they will not be good. Don't be listened to them. And that's actually exactly what I did, but unconsciously. I just couldn't bring myself to listen to it. And then I hesitated for another few weeks because I'm like, yeah, I don't know. It's not working. But I've already bought a mic. And then somebody wrote to me.

It was a designer from Kazakhstan. He found me on Instagram. And he invited me to be a guest on the live Instagram live to talk about design and product culture at Google. And I'm like, okay, so this sounds like an opportunity. So what we agreed with him is we did the live. And then during that live, we also recorded our audio to our computers. And then I made a podcast out of that live. I'm pretty good at Q&A, but I'm terrible at narrative.

And that became the first episode. It was also kind of meta. It was me talking about my experiences on my own show. But somebody else asked me questions. And that ended up pretty good. That's today's the most recent episode on my podcast. That's an interesting start. And I think that's one way to commit, right? Like throw lots of money into it. And you have to end up using it now. Going back when I was speaking in my first conference. And I always thought, I speak very well. I like it.

People tell me that I speak well and all. And then they put it up on YouTube like a month later. It was like, yeah. So I'm waiting for you, Ilia, to turn this into some magical voice. Yeah. Remember I did that presentation of the Voice Chatbot at the Invent in 2019? I think I actually never watched that video. I just couldn't bring myself to watch me talking on camera. I didn't understand something about it. And I just wanted to tell anybody who is aspiring here.

As part of podcast or as part of creator, you're not alone. If you're dreaded hearing your voice, you should dread seeing yourself on camera. You're not alone. Actually, that's what to make podcasting also easier than video. I actually had a YouTube channel where I was recording some music videos. And I was talking there, working on camera, especially if you don't like your voice in addition to that. Introducing so much more friction, so podcasting is lower friction experience.

Because you just talk. At least you don't have to be self aware about how you look like. It goes away with practice. I promise you. But the first two, three times, it will feel terrible. You just have to get through that. Yeah, and it's probably also like our own mental self image of how our voice sounds like and how we speak versus when you actually see it in third person. It's different. And that probably is a bit of a jarring experience.

But I'm already enjoying it. It's okay. We'll see how it turns out. Yeah, the thing is when you speak, you hear yourself speak through the inner ear as well as the outer ear. So there is some transmission of vibrations going on inside your skull and stuff. And when you listen to yourself on recording, you listen with your outer ear only. And that's why it objectively sounds different because you listen to it through a different listening device, so to speak.

This is about a year back, right? That's when you started the podcast. Do you remember what were the exact steps you took? How did you get guests for the next few episodes and all that? Or how did you promote the first episode? So the set I first got the mic, got that recording arranged, and then that recording set idle for about a couple of months because I couldn't bring myself to listen to that and edit it. Then I did edit it and I actually used a sound engineer for the first few episodes.

The same sound engineer that produced the podcast that I was a guest on. So you said not to bother, I just hired the same guy. We produced those episodes with him. I really wanted to do music for the first episode, but I couldn't find appropriate music. And I didn't have my instruments with me to record my own music. So I ended up actually releasing the first couple of episodes without music and the sound kind of dull. Investing music, music is great.

We produced the first episode, so I have this MP3 file. I created an account in anchor.fm, which was one of the first tools for podcasters for recording a podcast. I think actually their value prop was like you can record it on the iPhone and publish right away, which I think is a terrible idea. You have to use a good microphone to do good quality podcasting. So Spotify eventually bought anchor, so anchor is free.

And it allows you to upload a MP3 file, provide all the artwork, provide descriptions. And it will automatically publish it to platforms like Apple podcasts and Google podcasts and Castbox and Spotify. The one thing I'll say here right away, which I did, anchor or other tools like that could create accounts in Apple podcasts or Google podcasts for you. But you would rather invest your own time and create your own account linked in your personal email that you have full control over.

And then anchor will give you the syndication feed, the RSS feed, basically like a text file, which has the information about all of your episodes. And then you can just copy paste that URL into the Apple podcasts and Spotify and others. It takes more time. But if you have to move off anchor somewhere else, you are not locked in.

You would just have to like port your stuff, maybe create your own feed, but you don't lose, you don't have to like re-register those accounts because that's going to be terrible. You'll probably lose your following if you have to recreate that. So I guess what you're saying is anchor allows you to get started with one click or something like that, right? Where it'll create all these accounts and manage it all for you.

But what you're saying is go ahead, create your own Apple accounts and Google podcasts accounts and all that and then connect that into anchor. You actually connect anchor to those accounts through that feed URL, which basically just gives the Apple podcast knows, okay, so this is the feed that gets occasionally updated. And then Apple periodically checks for updates and if it sees a new podcast, we knew episode there, it will get published automatically.

But still, anchor serves is that storage source of truth for your podcast that other platforms get information from. There's no direct coupling, there is loose coupling through that URL, which you can always change if you want to. Actually had a problem when the Russia started to fall under sanctions earlier this year.

One of the changes that happened is anchors stopped supporting some of the Russian platforms, listening to podcasts and actually had to recreate a feed specifically for those platforms, not being reliant on anchor directly allowed me to spend some work and create this kind of parallel feed that helps me publish those platforms because no less. It makes sense, right, like your primary listener base is there, so you have to figure out a way to produce it in there.

That was the publishing part, which took maybe couple of weeks because it takes time to get approved. One thing I would have done differently now is probably I would have published a teaser episode just to kind of oily machine, so to speak, so that all the accounts get approved and it's already because when you publish a second, the third episode, it gets to all of those platforms within 24 hours or so.

But when you do the very first one, because of all the approvals that I required, it might take a couple of weeks between let's say Apple Podcasts and Google Podcasts. And then if you advertise it, it's not available on some platforms, it just makes for a weird promotion experience. And speaking about promotions, I already had an audience of about maybe five or six thousand people on Instagram where I just published like, hey, I have this new podcast.

I got maybe three, four hundred listens within the first week or so, so that's not the luxury that many aspiring podcasters will have, because if you don't have a platform to advertise your stuff on, then it just makes it more challenging. Actually, I'm curious how it will work out for us for this podcast because we don't have any English speaking audience. I mean, we do have our friends, but it's not like thousands of people, so you'll have to figure out how to promote this one.

And then I asked about sourcing guests. Here is a trick for you. My first few guests were my close friends. And then I also had a bunch of people who are secondary connections. And it was very helpful to have a safe environment for the first few episodes where you don't have to be self conscious about what people think about you if they have some extra time.

And these people, they don't care about being on the podcast. They just want to help you because they are your friends. In some of those people, they bought microphones because they just wanted to help me so badly. They even went through this hassle. That was pretty cool. I would imagine like, if I had Tim Ferris on my show as a guest, somebody who is like a professional would be intimidating.

Yeah, it would be intimidating, right? And then if I would be stumbling or speaking the same thing twice, that would feel a embarrassing. And Tim Ferris would probably be annoyed. He's a nice guy to know, but you would probably want to speak to those high profile people at a later stage and reserve the first few episodes for people who will support you and go out of their way to make you feel good about your recording. What were some of the best and worst experiences?

The best experience I had was with a friend of mine. She is really good at public speaking. And I did almost no edits to what she was saying. Because she had clean speech and also some of the things that she was talking about. She presented to like the large audience a week before at the conference.

So she had some of that really nailed for the offline event. And she was also very engaging. People love her. She also gave me a lot of reach because people were listening to my podcast because of her. Yeah, so we actually did two episodes of her on different topics, because she's absolutely fantastic. Her name is Yana. And the worst episode was I think these guests will not be offended because he knows what it is.

He's my old friend. We've known each other since middle school. So for about 25 years or so, we've been friends. And I was back in my hometown. We recorded an episode sitting next to each other in my parents' house. And some reason he was very nervous. He's like a great talker, very fast talker, very engaging. But he was very intimidated by the mic for some reason. And he couldn't quite get over it. And I was just trying to calm him down, taking some breaks, giving him some tips.

Eventually, he got more comfortable. I think I also asked him to not look at the mic. Look at me when you're talking. And then he got more comfortable. And the second part of the episode is much more natural. The first part feels choppy at times. But the funny thing is, after I edited the episode, I sent it to him. He was like, oh my god. That sounds so good. I expected much, much worse.

I would say the end result was a 100% improvement. 100% as a 2x improvement over what the actual experience was. Another not so great experience I had also was with my old-time friend. He wasn't a little bit of a down mood when we recorded. And it was hard to extract from him good long answers. Because I liked long form. I liked to just give it a person a prompt and let him talk for five minutes.

But the human was like, he would just say, yeah. And then you have to unpack that. And it wasn't as natural to me. And I had to adjust his style. And at the end, actually it ended up being a pretty good episode. But it ended up being a 30 minutes episode. And what was interesting is that people wrote comments. They were saying, oh my god, this is so cool. He's so interesting. His experience is so great. Too bad it ended too soon.

So that was just an interesting experience for me that maybe I should have more questions just in case. Because I thought we would just talk about two or three topics. And that would be it. But the conversation just didn't flow. And I think that's always a wild card when you talk to people who don't know. I mean, I knew this guy, but just he wasn't in the right mood. But with the people who don't know, you never know what it ends up.

So would you in these kind of situations, would you re record with them? My immediate reaction is probably not. Unless, I guess I'm still second guessing if it's his mood or maybe it's also a bit of part of his personality. I don't know if it would have been very different the second time. He actually might be listening to this. It's none of his fault. It's just there was a bit of disconnecting expectations.

Not expectations actually. Disconnected with my assumptions about how that would go and it didn't go the way I wanted. It's to turn out great. But yeah, I guess some of the things that I dread before recording episode of someone and it's really, really bad. Like, how do you go back to the guest and say, well, you know what? You know, it's one thing to say, well, your audio quality was bad. It's not really publishable. It's one thing.

The other thing is your story wasn't really exciting. This could be really difficult, I think. If you talk to somebody who you don't know, and they agreed to have a preliminary discussion with you to prepare, don't commit until you actually talk to them and figure out that their story actually fits into your podcast and also the vibe fits.

Then you don't also commit to like actually publishing that episode, right? After post production, you hear it and see whether it makes sense or not and go from there. I guess you've got the implicitly committed to publishing. I've heard of podcasters where they've recorded, but decided this wouldn't be a good episode, so never published it. I guess that kind of comes after you have become an established podcaster more easily than before. Yeah, absolutely.

Quick segue. You also talked about music, right? And you kind of recorded a few episodes without music and it was very dull. So you are a hobby musician. Do you put in your own music? Where do you get music from? What do you look for when you're looking for music? I had this idea of recording my own music for my first podcast. The trend in podcasting now is that there is EDM, the electronic music, usually.

And it sounds kind of nice and powerful. And I actually don't know how to make that kind of music. I played guitar. I had a guitar at bass and I could produce some drums. But that would be like rock, metal, and whatever I tried to come up with, it all sounded too aggressive. I just didn't work.

So eventually the sound engineer I worked with found an interesting composition on audio jungle, which is a website where you can lessons music for advertisements, for podcasting, for videos, for everything. Because my podcast is called IT Samurai, it has this kind of Japanese twist in the name and also in I guess in spirit. He found a dubstep music with Japanese theme, but he lessons that and use that. For Metacast I also found a nice sounding EDM music. So the very first episode has music.

When you say you license it, how much does that cost? It depends on the price that authors of your music said on the artist. The one for this podcast cost 120 bucks. For Metacast. For Metacast, yeah. It's a perpetual license and it allows unlimited distribution on podcasting platforms, YouTube, or all this stuff. There are also cheaper options. You can license it for 30 bucks, I think, or even 20 bucks. But it only gives you, I think, a thousand, maybe 10,000 of downloadable copies.

So I decided that it's not a big difference. Let's just go with the... Because we want to go a big go home, right? Let's just pay hundreds of 20 bucks. Let's committed. I mean, we are already home, so we only have to go big now. Let's talk about the process of you producing the podcast. I was surprised to find out how much post production work goes into all of this. How long would you say it takes you to produce an episode? How long are your episodes typically like an hour or so?

My episodes typically are from one hour to one hour, 20 minutes range. There are some short or some longer, so I did some time sheets. When I did my first two, three episodes, I'm like, it takes so much time. It feels like a second job to produce one episode in a month. So I did a time sheet. So you decided to add on to that work and collect metrics about it?

You know, if you can measure it, you can manage it. I did a time sheet, just like a simple Google sheets, where I just put approximately a number of minutes I spent on each stage and the date. The first few episodes were about 20 hours of time each, which included the time to reach out to the person, give them instructions. For some people, I had a pre-call, so we could discuss what we were going to talk about.

So I can do a more about them, especially for the secondary connections, who I didn't know as well. Then they actually recorded a couple of hours, and then about 10, 12 hours of post-production. Yeah. Wow. Because you get about two hours of audio, then you cut out about 40 minutes out of it to get one hour or 15 minutes episode. And that's hundreds of micro-edits that go in there.

And then you also have to listen to the entire thing, to make sure that it sounds great, and then sometimes you have to listen to it again. One important thing that I learned from a sound engineer is that you really want to listen to it in different conditions that people will be listening. So I would export an MP3, and I would listen to it in my AirPods, I would go to the car, I would listen in the car, I would listen in the car while driving. I would listen on the Alexa via Bluetooth.

You know what happens is if you mess up some frequencies, then the sound starts to sound really bad in some conditions, like in the car for example. It's not bright enough, if you don't have enough high frequencies, it just kind of blends with the trumple from the car itself. So you have to really listen in different environments. One thing that you said it's very curious is, let's say that you did find that it doesn't sound great in the car. What would you do about that?

So I tried to figure out why, and then I go back to my editor, I worked with the equalizer, good equalizer was the tool that let's you edit frequencies. You know, I have hypothesis, so hypothesis could be that there isn't enough high frequency. Add some high frequencies, I make it sound a little bit brighter, then I export it again, copy to my phone, go to the car, listen to it.

It gets easier with practice, especially for my own voice, I use the exact same setup every time, and I have this template that already has only filters, only effects. So my voice always sounds the same, but for my guests, especially depending on what kind of equipment they use, sometimes gets pretty tricky.

Actually had one guest, unfortunately we couldn't publish the episode, he was a CAO of Major Bank, and his PR gave approval for publishing the day before the whole thing within Russian Ukraine started. It was just inappropriate to publish the episode, you've never published it. That guy, cause of his environment, his audio was very bad. I did probably like 10 different takes on the exports to make it sound fine in AirPods. It was like booming, and I couldn't figure out what was wrong with that.

But there's also one of those things that gets better with practice, as you practice you start to hear better, and then it takes fewer attempts to make it sound great. I have heard some podcasts that didn't sound great, and it's very off-putting, so I do see the need to put in so many hours and work into it. This is the 10 to 12 hours of post-production you said, that's with all the AI tools and everything that you use, right?

That automatically remove clicks, and like my dog bark once while we were speaking, all of that, it automatically removes, and even with that it's 10 to 12 hours. So that was before I started using those tools. That was when I edit it manually, fully manually, and also send it to the sound engineer, and what I have in the sound engineer is, he would remove the breaths and other things that are unnecessary, or cut some pauses, and then he would send it back to me.

And what happened is, I would listen to that, and I would have a few feedback items for him. So send it back to him, he would make the changes, send it back to me. So in that 12 hours, there was a lot of inefficiency of re-listening to provide feedback. And this is where it's very important, if you really want to work with sound engineer, you need to find someone who shares your idea of what proper speech pattern is.

Because if there is a bit of a disconnect, and it's not the fault of the sound engineer, right? I don't blame him or anything. We just maybe perceive speech a little bit differently, and therefore our perception is different, because it could be subjective. I figured that it's more efficient for me to do more work. And for someone you said you didn't like the way your voice sounded. This is, I guess, torture, right? Listening to your own thing for hours and hours again and again to improve it.

But I guess that's how you get used to it and get better at it too. I guess one important thing to accept, that's how everybody hears your voice. Everybody, absolutely everybody, you just have to accept it. And it is what it is. An interesting thing for me was I never liked my voice, but when it first episode was produced, and people were saying that I have a pleasant voice, and I'm like, seriously? And then I listened it again, but that feedback that they gave me.

I'm like, I don't know, maybe, yeah, yeah. And then you just start to get used to it. And also, I think as confidence grows, you start to speak better, enunciate better. It's like in any craft or sports, you just listen, you adjust, you try new things. One thing that really helped me is actually did vocal lessons five years ago, which also helped me to get used to my sound a little bit. But you know, the singing voice sounds different from your speaking voice, but still it was helpful.

And also I did some vocal and speech lessons last year. And she gave me some tips and exercises for how to warm up for the recording. We also worked on enunciation. Actually, I want to resume that because that was super helpful. There is a timbre of your voice. You can't really change it, but this speech pattern you can. It just takes a lot of practice. How do you source your guests? You said in the beginning you interviewed a lot of friends, and that helped you, right?

Get used to the setup and all that. And so now how do you source guests and how do you prep for the interviews? I never had to do a cold reach out in a space like technology. It's a fairly small industry. And people have friends or their colleagues. For example, I interviewed one guy who was a second degree connection of mine. And he wrote back to me a few months later and said, I have another guy who might be a good fit for your podcast.

And then back when you know, both of us were still at Amazon, I was actually talking to somebody who also was at Amazon just for like a coffee chat or something. Three or four years later, this guy became the CIO of that bank that I mentioned. So I just reached out to him and LinkedIn. I'm like, hey, I have a podcast. Do you want to come over and join? So I would say I haven't exhausted my second and third degree connections yet.

Actually, I did reach out to a few folks who were like high profile people. I think I reached out to two of them. And both of them said no. And that's okay. I guess my podcast isn't high profile enough for them yet. But the fact that you're not actually doing cold outreach and yet people are going out of their way because they like the podcast or they like the experience of talking to you being a guest on it.

They're going out of their way to like reach out to people and letting them know about it. That's awesome, right? Yeah. That's the best way to get them. And actually, very last episode that we recorded. So that person he was subscribed to my blog and he would comment on some of my stuff. We started discussions, started chatting and then we eventually did the podcast together.

So that's another way to source guests. This person essentially became a first degree connection for me. But he came because I had a podcast and because I had a blog. So it's this kind of serendipitous cycle, I would say when you have a podcast, when they have a blog, so you invite people, they source more people for you. People listen to your show. They write back to you, start talking to them. Some of those are interesting enough that you actually want to invite them to your show.

I think that's just like a snowball that takes time to grow. But it does grow over time if you nurture that. And that's one of my challenges with podcasting. Podcasting is very decentralized. You have the Apple podcast, you have overcast, Google podcasts, YouTube, by the way. Spotify, I think only cast box and YouTube have comments. By the way, I get most of my listens from Apple podcasts. They don't have any way to do comments.

You mean on an episode because Apple has like ratings and reviews, but that's for the whole show. Yeah, but you can respond to them. What is missing from that whole ecosystem? For me, the creator is an ability to respond to the listener who left a comment and engage in a conversation and build it to relationship.

So that relationship building is basically offloaded to another platform to Instagram or Telegram. For this podcast, I think we should maybe have LinkedIn page, we should have a Twitter account. So that there is some venue for people to reach out to us because otherwise it's like broadcasting. I don't like broadcasting. I like the relationship.

You like the feedback and everything. Yeah. So how do you record? So when we first got started, I recorded to my computer in a digital audio workstation, DAW. That's what people in music call is. I use a reaper. You can use it for free. I have a small business license like 60 bucks. If you ever heard about Pro Tools or Logic, those are the higher end, very expensive software for music production.

A reaper is cheaper equivalent, which has a lot of the same features. My guests, though, they recorded on their phones in a sock. What? Yeah. Actually, that's how I recorded the podcast when I was a guest. The iPhone has a pretty decent microphone. It records pretty good quality. And there is an app called, I think it's called the Voice Recorder Pro or something like that. It's a free app.

And it records in a lossless format, in Wave, and there are a bunch of kind of knobs you can turn there. And the sock just serves as a pop filter. So I don't know if you ever saw podcasts, there's a video or the YouTubers or musicians who sing in the microphone. And there is a round mesh disk between them and the mic. So that's called the pop filter. It helps contain the breathing and some of those kind of speeds and they put sounds.

So that's why it's called a pop filter. So the sock actually serves as a pop filter for the recording. You can go pretty cheap on the first ones. We are recording this particular episode in the tool called, quadcast, squadcast.fm. I learned about it when I listened to the podcast called podcast junkies. Just kind of similar to our podcast. The guy has been doing podcasting and doing podcasters for over 10 years. And one of his guests was a person who produces the teamfares show.

It's a big deal, I guess, because he's podcast one of the top podcasts and they use quadcast for recording. I'm like, okay, so if teamfares uses quadcast, it must be good enough for me to also try riverside.fm. So you can do video and I will be recording in there. I think it works fine. I just like this quadcast interface a little bit better. So that's why we stick to that. It makes it a lot easier to record the podcast because you and I basically just talk over video.

Just like Google meet or zoom or anything like that. And in the meantime, quadcast in our browsers records our audio tracks in a lossless format. This will be a huge files. It will be about 500 megabytes per hour of recording. And then they will be uploaded to the cloud. After our session is over, I will download them and can do post production those files.

Whereas if you would be recording on your own computer or on your phone in the sock, we would have to find a way to upload this to a draw box, a Google drive. And I would have to download this. They will be out of sync. It's a lot of work. So I would say make your life simpler. And with the squadcast, I think you can use it for free for seven days. Riverside, you can record two hours for free. Just don't bother. Use one of those tools.

You were saying that in your previous podcast with riverside, you thought you lost somebody or something like that. Oh, that was terrible. We recorded in riverside because quadcast raised their prices. And I'm like, let's try riverside because it's cheaper. And so we tried riverside and the guy was in Armenia and he had a pretty slow internet connection, 128 kilobyte or maybe 256. It's kind of weird to talk about this kind of speed of slow because I grew up using modems with 14 kilobots.

That music is still in my head. Yeah, exactly. So when I say 256 is slow, it's like seriously. But yeah, anyway, so when you upload a 700 megabatt file, it is insufficient speed. And riverside didn't show any indication of the file being uploaded. So we actually thought we lost the file. We really freaked out because we were like, where is it? My track was fine because it was on my computer, but he wasn't. And he also couldn't download it on his computer.

So we had to open a ticket to support. We also tried to research if there was any local storage in the hard disk where these files are stored. And then while he was chatting with support, the file apparently finished uploading and became available. But it took maybe like an hour to get there. You know, I got some great hairs from that because if you do the same thing the second time, it's just not the same.

You know, I think somebody said you can't walk into the same river twice. Yeah, it would have been a different episode. No, I think even if we, you and I were chatting, this is very natural. I really like this the way we are chatting. But if you are going to do it again, I'm sure it would be way more artificial because we have already talked a little bit. That would influence what we're talking about. Yeah, you would have exhausted my arm up time. What? For this week.

We would have to wait another few weeks for this to sound natural because even within the week, I think would have not worked. What gear do you use? What microphone, headphone? I started with the Sure MV7 microphone that connects directly via USB. I use the Reaper Digital Audio Workstation. I use Quadcast. Right now I use the Bose headphones. I don't know the model, like $300 headphones that I bought from music many years ago.

I actually used to have studio headphones, audio technical ATX 50. But unfortunately, so we did a road trip in California earlier this summer and my rental car was broken into. We were moving from Seattle to Florida and we were in transit. We had a full trunk. A backpack was stolen. It had two of my laptops. That's studio microphone, my headphones, my iPad, my sons iPad, my Kindle, my car keys, a few thousand dollars worth of stuff.

I decided not to buy the same kind of equipment. I'm using my Bose headphones. I think they're good enough. For audio recording, I use a Sure SM58 microphone, which is a vocal mic. Pretty much any musician uses that microphone on stage. This microphone is really good at really picking what you speak into the microphone, which makes it a perfect mic for concerts.

Because it doesn't pick up the crowd or the instruments that play behind you. That's why you see these musicians who almost suck their microphone. It's because you have to be really close to it. This microphone is surprisingly good for podcasting. I think it sounds not too bright, but maybe my voice is bright enough. So it works fine with my voice.

I use that mic and I use a Focusrite 2i2 audio interface, which this microphone connects to and this interface digitizes the sound and puts it into the computer. This setup is actually more expensive than the other mic, because this audio interface is maybe 250 bucks. The microphone is 100 bucks. And also you have to understand some of that mechanics. So I would say if you're just getting started, I would recommend getting a USB mic because it's easier and cheaper.

So you use that and you ask me to buy a blue U.P. Why? You know, there is a part of me that actually regrets this recommendation a little bit. So I have a... I hope you didn't throw away the packaging so you could return it back to Amazon. I'm just kidding. Blue Yeti is a mic that is used by many, many podcasters and creators. It has tens of thousands of reviews. I actually have one of those. I use it daily for my video conferencing at work. It's a great microphone.

The challenge is that with all the default settings, it picks up too much noise. So you have to install additional software to reduce the gain. That kind of amplification that the microphone does by itself, then it works. To be fair, I haven't used any professional mics, but this... You're right, the amount of noise it picks up is crazy. But at the same time, this is, I think, honestly, the first time I've liked how my voice sounds after tuning all those knobs that you showed me.

So, yeah, I think Blue Yeti is a good starter mic. I think it was like $120. Yeah. A little bit more in Canadian dollars. We never actually mentioned where we are. I said that we moved to Florida. We are just north of Miami. But you're in Canada? I'm in Vancouver. Oh, by the way, talking about that SF thing, right? I have lived in Seattle for a while and Vancouver for the last four years or so.

And every time I go to SF and I'm with my friends, I leave my backpack in the car, right? And go to get a drink or whatever or an ice cream. And they're like, no, no, you're not leaving the backpack in the car. And I always get surprised why. But I do hear all these stories. And now, through you, like I'm hearing firsthand. One interesting thing that you said is they took the car keys also, but not the car. They didn't take the car.

No, it was a backpack. Backpack that had all the electronics and had my... Oh, your cars, not the rental car. Yeah. And I'm now recovering this second key because I'm worried what if I lose the only key I have left. It's like $500 for the smart key. And it also had stuff like the key ring that I really like. This whole thing felt a lot like a violation. It took me a while to actually get over it.

And luckily, it was the final leg of our trip. We drove from SF to LA and then back to the Sequoias. It was a two and a half week drive. And then SF was the last leg. I remember we entered San Francisco. I'm like, oh my god, I really liked it. I wouldn't want to leave there again because, you know, it's a great place for families with little children. I'll just leave it there. Some people would disagree, but okay, yeah. Yeah. Everybody has their, yeah.

Yeah. I wouldn't leave there, too, but I totally get what you're saying. Yeah. I like more space. I don't necessarily want to be in the city. I want to be in subgroups. But yeah, so it's like such a nice thing. And then before we checked into our Airbnb, we stopped by the ferry building to eat in this place called the Hogue Island Clim House. Clim Shower House. Something like that. Great clam shathers. I recommend everybody's.

Yeah. Now we are going very far from the Metacast, but okay, it's okay. Good. So yeah, we left the car there. And then we come back. It was a Toyota 4Runner, a large SUV with tinted windows and slurry of darkish outside. So I opened the trunk and then asking my wife, like, did you move the backpack? She's like, no, I'm like, where is the backpack? And then I noticed that there is a hole in the window. So they just smashed the window glass all over the trunk.

Luckily, they didn't take anything else. And luckily, I actually carried all of our documents with me, all the passports and all these kind of stuff. It was in a smaller backpack that I always carry with me, just because of potential risk like that. Yeah, anyway, so then we did like a police report and the rental company replaced the car for us. But last three or four days that we were in San Francisco, we weren't in the mood of vacationing.

We were just like waiting for our flight. That's what it was. Anyway, it's a detour. Long detour, but we started from what equipment you use and got all the way to recommending our restaurant with clam shoulders. So that's awesome. So now you've recorded it. You've put in all the hours. Now you're ready for post-production. What's next? Actually, I think one thing we didn't talk about is the preparation. I just want to go back a little bit to that.

I think it's important to really prepare for the interview if you do an interview podcast. To really get to know your guest. Ideally, you talk to them maybe for 30 to 60 minutes to understand the key interesting parts of their story. Try to extract from them as much as possible upfront. And then what they usually do is I create a Google Doc, which is shared with the guest in kind of edit mode.

They can edit it. I can edit it. I put the talking points and some of the stories as reminders for them. I try not to put specific questions. I also asked them to try not to like type what they want to say. Because if people read, it starts to feel unnatural. So what you really want to do is to have like an rough outline and the key things you want to focus on.

But then you can talk about clamshells. If that's where it goes, right? Because I guess the feel I want to have from the podcast is two friends talking and somebody else picking in on the conversation and conversation aren't linear. They never are in real life. So yeah, that kind of prep is very important. Coming back to your question about we recorded now. What?

Then I take those files from the squad cast, import them into Reaper. And I start cutting things out because at that point it's all about removing things. Here are some dirty secrets of clean sound. Our brains work such that when we hear people talk, we filter out all the starters, all of the filler words, unless they're super excessive.

Actually, I remember listening to, I think it was a podcast. Long pauses and filler words are human ways to basically indicate thinking progress to the person they're talking to. It's the progress in that dot, dot, dot spinners. Yeah, exactly. And the challenge with audio is that you actually don't see the person and you start feeling those long pauses. You start hearing those filler words, you start feeling those arms and ooms.

I actually cut those manually. I listen to the thing and as I go, I cut out those things. Sometimes I have to leave them in because if you cut them out, it starts to sound unnatural. It's okay to have a few filler words here and there. And actually, there are some people who are really bad with filler words, including myself, to be honest, when I speak in Russian, I think in English, I have fewer filler words, maybe because I'm more self conscious.

There are some parts of the recording that just don't work and I cut them out entirely. And then you have to watch out for things like, as I said before, and if you actually cut out that thing that the person said before, you have to work around that. I actually had a situation where I had to leave the whole thing I want to cut because there was like path dependence where later content dependent on the previous content.

So we can't remove clams out there from this episode now. That's what you're saying. I don't think we can. I know some people record some parts that they messed up later on. I don't like that because it's just, I don't know, it's too much work to start with. And it also to me, it feels very unnatural even when you're hearing it. You can almost always pick up that this segment was done in a completely different context.

Yeah. I think it depends very much on what your format is because if you do an interview, you want to sound natural. If you are NPR and you record half an hour shows like in Visibili that heavily produced with interviews and music, there is more opportunity for this kind of post editing and additional recordings. And they also have to sound really, really nice. Like we don't have to sound as nice. Yeah, I cut those out.

And then there are artifacts that the microphone speak up that you wouldn't be even aware of if it wasn't for the good microphone. Things like breath. When we zoom, for example, it actually filters out those breaths. So we almost feel like the people around us don't breathe. We don't even think about that. But on the mics, you hear this. We will now play in audio clip that actually shows the software isotope RX9 and their D breath filter that filters out the breath.

And sometimes it's funny to hear how different people have different breathing patterns. And sometimes you breathe really often, some of you breathe less frequently. And it's so funny because you almost lose people's voice doctor because you know more about their voice than they know themselves. Another funny thing is mouth clicks. Most of the people who had been on my podcast, including myself, we are very heavy mouth clickers.

It may sound unpleasant, but there is saliva on the contrary, your mouth could be dry. There are lots of small micro clicks that you don't perceive when you talk to a person. And actually you wouldn't perceive when you listen either. But when you remove them with software, again, I use isotope, it just, you know, it's like those small things they compound and they make for better output audio quality.

And now I'll play an audio clip of just the mouth clicks that were removed from the previous few sentences. Another common challenge with recording is echo. Most of the rooms have what's called reverberation. So basically sound bouncing off the walls and doors, windows coming back into your mic. So you say something and then milliseconds later the mic picks up the reflection of that sound.

So with echo, what you usually try to do is either sit in closets like you and I do right now because we have a lot of clothing around us and that clothing absorbs the sound. So it does reflect back. Or even if you don't have a closet or it's not appropriate for podcasting, you can put something behind you. Very thick curtains work really well.

Or bookshelf with books because books actually tend to absorb sound as well. When I was recording in my parents' house I put cushions from the couch that just stole from downstairs. And then also my sound engineer calls it a pillow house. If you have a mic that stands on the desk, you would build a house like a big one around the microphone. So that helps absorb sound that sound.

Like a could be removed with isotope or other tools like that. But when you remove something from audio, it reduces the quality. And audio production is very much garbage in, garbage out. So if you have poor quality in, if you remove too many things, it becomes even worse. So we always try to fix problems at the source. Sometimes I do some slight echo removal when needed.

One of the preparation steps is have your guest do their setup ahead of time so that they can record test audio, send it to you and you can give them tips to adjust their setup if needed. Talking about the echoes one thing you and I were talking about yesterday is I'm inside the closet and my first setup was facing the door. The closet is a small closet. There's only one door and I was going to face that and you mentioned that that's going to create a lot of echoes.

This is something I would not have even thought about that I should not be facing the door and that's like a whole big surface of polished material that's going to echo the sound quite a lot. And the other way around instead. Yeah, your audio sounds great. And I'll listen to that on loud volume later on. If there is a bit of echo, I'll do a slight adjustment. Yeah, let me actually play an audio clip that I recorded in a very echoey room and then how it sounds when you remove the echo.

I'm recording this in a room that has very high stillings. It's very spacious room. So there is a lot of eviboration. And to make things even worse, there is also a washing machine working in the background that this blue Yeti mic should be picking up. And now we turn on the echo suppression plugin in Isotov RX9. As you can hear, it starts to sound more like Zoom. It doesn't sound natural, right? This blue Yeti mic that you said, right?

When I'm just using it locally and I can hear all the background noise and versus when I'm on Zoom or you and I were talking on Google Meet the other day. And it takes out all of that. I was surprised, but it does a really good job taking out all of that. But then it does. You do lose a lot of the timber and the natural tone of your sound and all that along with that.

And the last thing that we need to pay attention to is the noises. Because even if you're in the closet and there is no sound reflection, if you have an air conditioner working nearby or a dishwasher or a washing machine in the next room that low level rumble, the microphone will pick it up.

Again, there is software in Isotov. There is a plugin that helps you remove that. But it also comes at the cost of reduced audio quality. And I'll now play a clip of me recording in the bathroom with the fan on. So you can really hear the noise and how it gets removed and what the end result of that is. I'm recording this in a blue rating mic in the bathroom where I have fan working. And also there are some TV noises and dog parks coming from outside.

And now I'm turning on the noise reduction filter in Isotov RX9. And I'll also turn on the D-reger filter. So we also use the echo. And you could hear how terrible the end result will sound. Yesterday I had the dryer on while I was chatting with you. And the same thing. It was like you can really feel it when you hear through these headphones with this kind of a mic. How much of that noise it picks up?

Yeah. And also little children dogs. When I did my very first episode, my wife was in the back of the school children in the kitchen, two rooms away from me. And we actually could hear some of the dishes in the recording. And just as you said dogs, my dog barked. I don't know if you heard it or not. I didn't. We'll see.

Okay, so that's how you set up. You've got like everything buttoned down to the last detail. What do you do about your guests? How do you make sure that they have the best quality? I was lucky with half of my guests who got microphones. They either had microphones or they purchased microphones. Actually, one of the people we recorded in the same room, which was really cool, for those who don't have microphones, what we usually resorted to is an iPhone in a sock.

Believe it or not, we couldn't figure out anything better than that. Because if you record through regular headphones or the microphone in the laptop, it's not optimized for high quality audio recording. It's optimized for maybe it's a bit of a detour into the MP3 standard. What MP3 does is it makes your audio 10 times smaller than the lost less wave file.

And the way it does it is by removing frequencies that are considered inaudible to the ear. And that's exactly what those non-professional microphones do. Their job is to transmit the voice as clear as possible with as little bandwidth as possible. So they aggressively cut out things that aren't considered speech. But our voices are much richer than what they think voice is. Have you ever tried recording with like AirPods? How does that work?

One of the worst things you can do for your podcast is to record them with AirPods. Especially if you have somebody who is doing something and walking around the house and then you get really poor audio quality. No, no, you have no. It's just so bad. And I don't think you can really fix that quality with any software. And I guess if you get a high profile guest, I don't know, like a president of the United States and they only have AirPods.

I guess if you go with what you have, right? But if you can, you should at all costs avoid AirPods or all those other lower quality mics. I love my AirPods, right? But I guess I'll just keep using them to listen to podcasts while walking my dog and not to record podcasts yet. I guess the irony here is that most people who have iPhones probably listen to their podcasts through their AirPods or in their cars.

I guess cars in AirPods are probably the most common use cases here. So that's an essential listener device, but an anti-patterant for the recording. The caveat I would give here is that my AirPods are non-pro because I don't like the shape of the Pro. Use the regular AirPods, the second generation, which I bought two years ago. I don't know if the new AirPods have a better mic. You can try it out.

By the way, talking about that, the two AirPods generations, the first ones I could not even use it. Like my ears would pain after 10 minutes of putting them on. But the second one that I got, which they're very weird with their numbering. So I don't know if it's the number three or number two or whatever, but the latest, through all one that they have, those are so good. I can put them on for hours and hours and no problems.

One thing that might make our listeners self-conscious, I'll say, is I was always amazed how you look at people who wear AirPods and they stick to their AirPods. And they stick out of their ears in different ways. So somebody had like, they stick to the sides, somebody like, stick to the front. It's because this shape of the ear is ear bowl, is it what it's called? It's different. And like in your case, they don't even fit.

In my case, they fit perfectly. They look kind of organic. Some people, they stick out, looks kind of weird. But the thing is, they still work. So Apple was able to come up with the shape that really fits, I guess, a large percentage of people's ears, which is just absolutely amazing. And the new ones, doesn't it? Like when you put them on the first time, there's a weight basically understands the shape of your ear and molds to it. And there's a mold to set that up the first time you use it.

Oh yeah. So maybe I should try the new AirPods Pro's. So we are about a hour and a half in, are there any other post-production techniques or tools that you want to talk about? No, but talking about power dependency, to the listener, that will be probably like one hour in. Because the first production, this is so meta. That's why you're listening to the meta cards.

All right. I think you mentioned already the equalizers. You listen to your car and if it sounds very different than you'll play with that. And otherwise you leave them all along. Oh right. Yeah. The equalizers. There are two important things I forgot to mention.

Yeah. The equalizer is basically tool to edit the loudness of certain frequencies. So you can amplify certain frequencies. You can lower certain frequencies. It sounds kind of scientific, I guess. But in every person's voice, there are certain, I don't know if overtones, undertones or whatever they call, I forgot the term. But they kind of sound unpleasant. And so you want to get a free move, some of that. That somewhere like in the low to middle range.

The tip that I got from one of the videos that I really liked, you just get that equalizer, plugin. And then you amplify certain frequency. And when you listen to it, if amplified it sounds terrible, it's most likely that weak spot in that person's voice. And then you slightly lower that frequency down so that you hear less of that. Some of the non scientific terms that people use for this is like it sounds boxy, like in box.

And what I do is I basically watch the few videos and the read a few articles. I copied the patterns that these people do, which is like cut out all of the low end so that there is no rumble that's coming from microphone, the background rumble. And then you also amplify the high end little bit so the voice sounds brighter. More kind of cuts through. Let's see if you're driving a car or if you walk outside with your dog, there was somewhere around noise.

Brighter sound makes it easier for you to hear the voice. And then in the middle end, I just do what they said. So it's like bring it up, see if it sounds terrible. If it is bring it a little bit down. So the second thing is compressing the sound as we speak our volume varies. Some things I can say really quietly. And sometimes I can really go loud, especially when you love, for example, it's very loud.

If you turn your head, it becomes a little bit quieter. Yeah, I've noticed myself like sometimes stepping back away from the microphone, but sometimes being close to it. Yeah. So sometimes back, they invented a sound compression from music, which basically amplifies certain parts of the audio.

So that basically you have, you'll say you have that song and there are certain parts, maybe acoustic guitar and the person singing and then you have like the drums and electric guitars and everything comes in. So if you just kept it as is, that would be like jarring because the volume job would be super incredible. What compressing does is it makes that acoustic part louder while keeping the same dynamics of that. So it still feels like soft and mellow, but it's as loud as the electric part.

So the same thing compression does with the voice. It basically evens out those volume deep send rises and that's pretty complex plugin to figure out. There are like a whole courses on compression for podcasting. So if you want to do your own sound production, it might make sense to watch those videos and experiment. This almost sounds perfect for an AI like machine learning based product to solve, right? Yes. And actually I need to explore if there are machine learning plugins there.

You and I are recording independently, right? In Vancouver and in Miami, very far away. So we're going to produce our independent tracks and you'll be able to have different levels of compression on both of those independently and then finally stitched that together, right? I will apply effects on each of the tracks, breathing reduction and equalizing equalizing depends on the voice, right? I actually need to double check. I think I play compression for both tracks at the same time.

So we both sound at the same volume because it could also be very jarring if your volume is different from mine. So it has to be consistent across the entire thing. If in a world where things are back to normal, well things are slowly getting back to normal and you and I are in the same room recording here in Miami, by the way, things are back to normal. You're in Florida. Yeah. It was normal like in 2019. But anyway, we won't get into that.

So you and I are, let's say we are in the same room and we're recording with the same mic. So you're producing like one stream for both of our voices. Would it be more difficult to do these like compression and equalizers or would you still be able to figure out how to independently split them up and do that? Yeah. So the ideal scenario there is that you use two mics and then there are certain tools that control for the bleed. I think it's called the bleed over.

When your mic picks what I'm saying, there are ways to control for that. So that will be the ideal scenario. If you record to the same microphone, which I did with one of my episodes, the biggest challenge is that when people get excited and they interrupt each other, not necessarily like in a bad way, but like, oh yeah, yeah. You know, all like the love, for example, in the background. Then you start to mix voices on the same track. Those are pretty hard to edit.

I would split the people into different tracks and I'll edit them separately. It's a bit of extra work, but it's totally worth it. And those parts where there's overlap again, we tried to control the source, but if it happens, well, we just still did it. So it's been a great episode so far. I've learned so much. I've asked you so many things too. It does feel a bit intimidating though.

If I was going to do this myself, I wouldn't know where to start and how many things I need to do in post production and all that. Are there like turnkey kind of services available or is a sound engineer the best way to get this done if you're like a aspiring podcaster, but either don't have the interest or the means or the time to do all the post production yourself.

Yeah, you can always pay for post production. So there are podcast producing companies. You just give them the audio, they give you the output, they can even help you set up your podcast. They can do everything for you. And I believe there are some that aren't too expensive. Actually, I want to try some of those. What I want to do with this podcast is I want to produce a couple of episodes myself, so I set the call to bar. And then this could be the input for whoever you might hire for this.

You're producing the test data. Exactly. Only if humans learn as well as machine learning models. What advice do you have for aspiring podcasters? It depends on what you start from. You could be a really natural talker. Then the thing you have to figure out is how do you actually record this. And recording is very simple. Get a mic, get a squad cast or the digital audio workstation, just record it. Are a phone in a sock like you said? Oh, phone in a sock, right? Yeah.

There are lots of great articles about how do you set up your environment. But just like we were discussing with Arnav, going to your closet, surround yourself with clothing. With something soft. And yes, what did I mean? And recorded that's not that difficult. And then if you listen to long tale of podcast, you will hear a lot of podcasts with absolutely terrible audio quality.

And they still have their audience. So if you have something to say that people need, then I think even if your audio quality isn't super great, that will still work. I would say don't feel like it's oh my god, it's intimidating. It's so much stuff to learn. The reason why for me sound quality was so important because of my background with music. I just couldn't accept anything less than almost perfect. Yeah, knowing you, I know what you're saying.

And I know that's why the post production that you want to do yourself the first few times to set the quality standard and cool. Exactly. Yeah. But I have the skills, right? Well, at least I had 70% of the skills I needed for this because I didn't do so much voice before. But I knew the tools I understood they quality and audio in general. So that wasn't a heavy lift for me, but I can imagine for somebody who never done audio production, that's a heavy lift.

You could even do things like if you do interviews, you could just record them in zoom. I mean, they will not sound as great, but our brains are really great at filtering out noises and echoes. Find some recording of a zoom call on YouTube and listen to it in good headphones. You will notice that after five minutes, you will stop paying attention to the echo.

Yeah. And you're actually listening to that content. So I think at the end of the day, it's all about the content. Do you have something to give to the world? And audio production you can figure out incrementally. That's a cool thing about the podcast. You don't have to do everything right away. Even the music, I was really stuck on music when I got started and I just decided to go without it and it still worked. As I said, it's the most listened episode, either without music.

Maybe I'm just being too much of a perfectionist with that because I feel like it's dull. But people who listen to it, they come there not from the music. I think music is like great for recognizing the ability of your podcast. It keeps it engaging too, but like you said, you can add layers and layers of this over time. Yeah, and if you're cool thing about the podcast is that you can replace the old file with the new file, unlike let's say YouTube.

So you can always add that music later if you wish to do so. That was the advice if you're a natural speaker who doesn't have skills for audio production. If you are not a great speaker, but you feel like you have that urge to do a podcast, it's important to figure out what is the mode that works for you. Like, can you train yourself to narrate things or like are you better at Q&A question answer like I am, maybe there are some other things.

Actually, maybe podcast is not the right format for you. Maybe video where you share your screen and click things around some kind of like a technical thing, right? Maybe the right format for you. I don't know what it is, but my point is find some format where it feels easier and then you can build up from there. Actually, speaking of narration episodes, I actually did one episode where I narrated and it took me a while to record, but I did it.

It was my, I think, eighth episode when I already got that confidence of speaking to the mic. At the end of the day, it all comes down to your confidence level. And if you like that in beginning, create a comfort zone for yourself. Make it easy for yourself, doing an episode with a friend. By all means, just try to not overstress yourself. I forgot who invented this concept, but it's like the concept of small steps. You don't just do one flight of stairs and one step.

You just go step by step and it feels much easier and you also feel what of progress is you're doing this. So basically, don't hold yourself back, right? Like experiment, create it, see what happens, go from there. You've got nothing to lose. Yeah. Recently, I was listening to an old episode of Team Ferris interviewing Neil Gaiman. Ferris is my favorite podcaster, Gaiman is my favorite writer. I think it was like from 2014 or so, a long time ago. And the audio quality is so bad.

There are two things. There's lots of breathing and if I'm not mistaken, there is also echo. I'm like, seriously? But then after a while, Gaiman had so many interesting things to say. I just stopped paying attention to the unimportant things like the audio quality. This is just to reiterate that as long as you have great content, everything else can be figured out later. I think a bit of a counterpoint here, okay? Because one of the podcasts that I used to listen to, the content was great.

I won't go into details of which one it is and all that. Now hearing you talk about what you do for post-production, it seems like they did not do anything. They would just basically upload it. So they would say things like, oh, let me take a sip of coffee and then actually take the sip of the coffee with the slurps sound and all that. Those things can be very distracting when you're listening to a podcast and you forget about what was being said before.

I'd say those things I'm assuming that those are very easy to cut it. They are. So at least do those minimal things. Yeah, that's true. And also, maybe it's important to figure out the vibe of your podcast. For example, what we've just done in the last one hour, 50 minutes, two friends talking about the topic that is interesting to both of us.

Yeah, we did a detour to a clam chowder and it's fine. But then there are podcasts where it's just so strictly professional that any kind of detour to a clam chowder would be weird. And because they are listeners, I can't do this. This is the half an hour of just content dance episode. And they don't accept anything else. Our show would probably attract people who are more into dropping into some of these conversations in a good way.

So this kind of live discussion, whereas there are people who like they value their time. They don't want to listen to half an hour of chitchat in addition to one hour of useful content. They would just try to find something that's compressed to all those people like go read the book. I bother without you at all. I can book some more efficient.

Anyway, so I mean every format has their own audience. I think it's important to be authentic because if you're not authentic, then you may end up just burning out because you do the thing that you don't like doing. Yeah, also I think while listening, you can very easily tell when somebody is authentic or not. Yeah, one of the things that I often get on my blog or the podcast is like, oh, can you tell about this? Can you tell about that?

And sometimes it's a topic that is aligned with my interest. And sometimes it's like, no, I could talk about that, but I just don't want to. And they just don't. And that's very important, I think, to have some kind of principles behind your podcast. Yeah, that's something that actually might be interesting to mention. When I started my first podcast, I defined a few principles. One of those principles was the experience should feel like fun for everybody.

Fun and enjoyable for me in first place because I don't want to be able to do this podcast. And for my guests, because they're invited, like I would never touch the controversial subjects or things that may put the guests in an uncomfortable situation. If I accidentally do, we would just cut those out. I wanted to be the highlight of their day. I want them to finish their recording and be excited for the rest of the day, right? Energized. And then I also wanted to be fun to do listener.

I want them to take something useful, but I also want them to just be delighted by the experience they get. Yeah, there's actually podcasts that I used to listen to. It's called this is product management. They really tight on their format, how they edit things. It's like very content dense half an hour, which sometimes is helpful when people talk about specific things like, for example, what metrics they measure.

But if they start to veer into more kind of soft skills kind of thing, it feels like it could have been a longer episode with more back and forth. But as I said, they probably have their own audience that values just that. What I really like is like Joe Rogan. But sometimes I look at these four hour episodes, I'm like seriously, it's like half an audiobook.

I would literally listen to Joe Rogan episode as if it wasn't audiobook because it's just so long or some of the teamfares two, three hours episodes. Oh my god, this is so long, but they're so good. I really like those. You know, you and I both listen to the legs podcast. I actually don't. Oh, you don't. Okay. I do. Those are like four or five hours. I don't listen to all of them, of course, because that's all I would be doing.

If I was listening, but but the episodes that I do, it's the subject that I find interesting. For example, astronomical biology or alien life or something like that, right? And those episodes, I find that it just flows through even though it's four or five hours long. One quick thing I would also add is the experience post production.

If you start doing podcasting and you let's say in the beginning, you just outsource it to somebody, right? And somebody does it for you. I would say at some point, take the time to do it yourself once because this is coming from my other hobby is like astronomy, right?

And at first, before I got into it, my interest was actually looking at these amazing things out in deep space that you can't see with your naked eyes. But then after about a year of doing it, what I figured out is that I really enjoy the technical process of setting up the telescope manually finding in the sky, like how to navigate in the sky, how to get to that object that you cannot see with your naked eyes.

And I just like point somewhere and look at it. And I started enjoying that process the technical parts of it a lot more to the point where now if I go out into a star party and set up my telescope, I look forward to that part almost a little bit more than the actual observing of the object in the deep sky.

You may find that you really enjoy the process of post production and playing with all these things and start doing that more and more. Yeah, actually, I think one thing that is very common to you and I maybe that's also why we get along very well is our craftsmanship in the things that we do. You're really craftsmanship software engineer.

I take my writing at work as a craft and I take actually podcast post production as a craft as well as interview if you have that mindset of craftsmanship in other parts of your life. I think what might be different with podcasting is that it might just seem to intimidating because like where do you start?

You just start with small things to this do that and then over time you grow your skills while you actually understand what you're doing, but I think understanding what you're doing is really key. You know, it's kind of similar to software engineers in some ways, these days you can write software by just writing glucose that combines multiple things together, right?

Without actually writing the things that actually do the work and the same thing with podcasting you can use all these tools and just like hack things together, publish it and just works. But if you really want to improve it, you have to go under the hood and use your own hands. So is there a stack of workflow for podcasting then?

For me to copy paste everything from the show notes, I will add a link to the podcast post production guide with Ripper, which is the digital workstation I mentioned, the cheap one with a bunch of re- and open source plugins that they use. If you just follow that guide that gives you a decent understanding of the process and also you can start with a template, then you can evolve into more things.

So for example, I actually, I learned from that post about keyboard shortcuts for Ripper and some of the concepts that go along with those. After using those for about a year, I was thinking maybe I should use the keyboard more because I started to get repeated stress in jury on my hand from using the trackpad and mouse. I bought an ergonomic mouse, doesn't help when you do hundreds of micro edits, the risk starts to hurt.

I created a bunch of shortcuts to optimize my workflow and these are the things that you do when you become more proficient. And there are probably even more things I could have done which I will figure out a year or two later. As any craft, I think it's important to just keep improving yourself. Not just in post production because post production might be too technical, not everybody on nerves like you and I, right?

But taking pride and approaching your interviewing skills like craftsman could be an interesting application. It's like how do you do storytelling? We didn't even like how do you actually make the other person to the storytelling? How do you put the guardrails? How do you move the conversation back from clam chowder's back into that topic? Exactly. And how do you actually finish the two hour recording and go back to work? So let's get over with it. Yeah. Last few things before we wrap this up.

For somebody who's just getting started, what are some of the resources you would recommend? For somebody who's just getting started, what are some of the resources you would recommend? There's a great book called the NPR's Podcast Startup Guide. We will also link it in the notes. It's a book written by the authors of NPR's Podcasts. They are probably the best people in podcasting, they're the gold standard. They've been doing radio broadcasting for, I don't know, like centuries.

I don't know how long they've been around. And then it's switched to podcasting. It's just a great book that gives you an overview and probably even simpler language than I was using in our discussion today. Great book. I highly recommend it. Then there is the podcast called Post Light. It's a podcast about something else. It's about technology. But their episode number 300 is a sort of anniversary episode where they invited their producer of the podcast to talk about the podcasting.

And that's a fascinating half an hour episode about how she works with them as a producer. What kind of speech impediments do you have? It's not like that. That she actually removes when I discovered post light. So this guy, Ritchie Ali, one of the co-founders of post light, his speech is so incredibly clean and articulate. And he non-seats so well. I'm like, oh my god, I want to be like Ritchie Ali in my podcast.

And at the same time, it's just so intimidating because that's not how I naturally speak. So when I listen to that episode and she's like, oh Ritchie, you stutter or like something like that. I forgot which exactly it was, right? And I'm like, oh, okay. So obviously he's a great speaker, but he is not perfect. And that's also very important to recognize for the podcasters. Actually, I should have said this as the first thing.

When you listen to things, you might get intimidated by always great speakers. But you don't know that they use so much post production that they actually might be absolutely terrible speaker. If you listen to them in real life. And what helps sometimes is you look up that person on YouTube when he speaks with video because with video you cannot do as much post production because you know, you can't easily chop the video. And you can hear how they actually speak and you might be surprised.

Coming back to that post light episode 300, it's very eye-opening about post production. And what is it that you don't hear in the podcast? Yeah, you recommended that episode to me last week, I think. And I would say even if you're not seriously thinking about starting a podcast, listen to that episode. It's hilarious. You learned so much in 30 minutes. Yeah, and just post light podcasts. I can't recommend highly enough, especially the first maybe 100 episodes.

I like when Region Paul just talked to each other. These guys have such great chemistry. And they are so different, which makes their conversations so fun to drop in. Okay, so let's wrap this up. Where can people find us? We haven't set up our substack or linked in or Twitter yet. Our website or podcast. Action now, now that's a meta cast, right? So you can find the links to our social media in the show notes. And we will register and add them after we record this episode.

So the ethical thing, right? So like we are talking about podcasts, so we can talk about these things. Yeah, we haven't done any of that yet. But we will before we publish the episode. And I think this is just like exactly what you were saying, right? It was what two weeks ago where you and I were chatting and thought, huh, why don't we do a podcast? And you said they buy this mic. And I think I got the mic on Thursday.

I was out all weekend. And so here we are the next day after the mic recording an episode. And it was a lot of fun. Yeah, it's Tuesday morning now. And as I said, you were out. I dropped some outline of the episode in the ocean on Sunday, I think. Yeah, now it's Tuesday and we already recording it. All right, so that's it for this episode. It was awesome. I learned so much. I think we'll try to follow the same kind of template when we talk to other guests.

It will be awesome to hear about like their stories and what kind of tips and tricks. And as well as like why they do the podcast. They do. If you like this episode and you want us to have your favorite pod custard over, please let us know. Again, we'll have our information and the show notes. And that's it. Stay tuned. We'll have more soon. Yeah, let us know and also let them know. Yeah, you would love it. So it's easier for us to reach out to them. Well, I guess bye-bye, listen to this.

All right, again, stop wondering. Sorry, now that you've done your first ever podcast episode, how do you feel? I thought it was very natural. I loved it, except you and me. Nobody else gets to see the video, thankfully, with the closet around us. But I was smiling all through, right? I absolutely loved it. Yeah, I was actually thinking while doing this that the same amount of information that we talked about today, how much effort would it be to write it out?

And you and I both were kind of meticulous in writing. So it would probably take me like a few days to, I would kind of play the whole scenarios in my head over and over again. Should I talk about this or that, whereas a conversation like this in a podcast form was very natural. It's not as structured but very natural and I feel like the learning or the information that came out would be equivalent. Yes, actually, I was thinking about that same idea when I was putting together the outline.

I'm like, this could have been actually a great long form post. But then I'm like, if you go into all the details, that would probably take tens of hours to really do well. And we actually eventually might wanna do that. If you just transcribe what you were talking about, that would be a mess. Like, clean childhood sprinkles all over. So that's what I actually like about the podcasting is that you don't really have to have defined path.

You have a direction and then just follow whatever it takes you. Podcasting is very inefficient compared to books or even the audio books for knowledge acquisition. But it's more interesting to listen to.