hello and welcome back You've probably seen the amazingly realistic AI conversations that Notebook LM can generate for free from any source But did you know that you can take those audio dialogues and convert them into a video podcast episode I've covered how to do that in various other videos However the methods are often timeconuming and involve multiple steps But I just learned about a new tool that does it all automatically And so I wanted to share it with you And I really mean automatically because it does the whole process for you It takes your notebook LM audio and converts it into a video podcast No need to manually separate the audio or anything like that It's an entirely done for you process Here's what it's going to look like We're going to generate an audio dialogue in Notebook LM bring that dialogue into jog AI which is the name of this tool and from there create our podcast video from the notebook LM audio and then we'll look at some other ways that Jog AI can generate talking avatar videos from pretty much any source This is just a two-step process and we can have it done in less than 10 minutes Okay so let's get this jam started So the first step as mentioned is generating an audio dialogue with notebook LM and our notebook is going to be on the topic of vibe coding So we're going to input two sources First off this video from YouTube from the AI daily brief and then this Wikipedia page also about vibe coding So we click create new to create a new notebook Click YouTube paste in the YouTube link that's now getting processed and then we can also add our Wikipedia page Okay so that is all processed and we can now generate our audio dialogue This takes a few minutes Okay so here it is Let's have a brief listen You know that feeling when a new wave of technology hits and you just sense things are about to fundamentally change Mhm We're diving head first into one of those waves today Something people are calling vibe coding Exactly Yeah We've been looking at some really fascinating sources describing it Vibe coding It's uh it seems to be a real paradigm shift in how software gets made Okay Okay Stunningly realistic output from notebook as always And now what we're going to do is download our audio and then we're going to move on to the next step We come over to the Jog AI website I've left the link for you in the description below so you can follow along Just click start creating for free You can just sign up with your Google account in a couple of clicks and you'll be in Now once you're in there are a lot of options here but what we want is the video podcast And so just click here and that will load Okay So as you see here at the top we have our sources panel and we have three primary options which are generate script upload script or upload audio We're going to choose upload audio We're just going to literally upload the notebook LM output exactly as it is and let Jog AI process it word for word The audio you upload has to be less than 30 minutes long which is fine for notebook because the conversations that notebook generates are in the region of 20 minutes So let's just click here You know that feeling when Okay And this has a duration of 19 minutes So that's within the limit And we click open Okay And that was quickly uploaded And we just hit the next button So that's generating our script And that's already done So now what we do is we choose our host avatars Now here we either choose these preset avatar pairs or we create a new pair And I'm going to create a new pair So I'm going to click create new Okay So let's first of all do avatar one I'm going to click here to select a pre-existing avatar Now it's important to note here because the first speaker in the audio is female then avatar one must also be female I think I'll choose this lady here Okay Now for avatar 2 let's try this guy And another thing to note here we don't need to select any voices because the point of this exercise is to use the original notebook LM voices So just ignore this section here because it actually makes no difference what you choose here The generated video is going to have the notebook voices Okay let's call this Jonathan's Duo And now we have to choose a theme I really like the mountain lodge I'm going to try that and generate the preview Okay And now we have to choose from one of these eight proposed pairs Or you can generate more like this Let's choose this one with a fireplace and click generate avatars Okay this will take four credits and submit All right that's an estimated 5 minutes ETA So let's come back in a second Okay our avatar pair is now ready and all that's left to do is to hit the render button and sit back and wait for the magic to happen And here's the final output Just before we go any further I think it's worth mentioning that this solution still needs perfecting but I do think it's interesting enough to show you This is a brand new feature and it's being constantly improved You know that feeling when a new wave of technology hits and you just sense things are about to fundamentally change Mhm We're diving head first into one of those ways today Something people are calling vibe coding Exactly Yeah We've been looking at some really fascinating sources describing it Vibe coding It's u it seems to be a real paradigm shift in how software gets made Okay So how does it work Think of it like using natural language you know just describing what you want to guide these really powerful AI models that are specifically tuned for generating code right so the traditional image you know the programmer hunched over endless lines of code that's changing it seems to be evolving yeah so that's the output it gives us do let me know what you think of that in the comments below I'd love to hear your thoughts just to say I ended up making three different avatar pairs and this is the one I liked the That's why the avatar's clothes are slightly different As mentioned I don't think this is a perfect technology yet but I know the Jog AI team are working constantly to improve it However it is certainly quick and easy and you don't have to worry about diorization or manual cutting between avatars All of that is taken care of by Jog AI Another big advantage of using Jog AI is that you can retain the notebook LM voices and original conversation which I know a lot of people want as the quality is the highest in the industry Having said that I want to show you another jog AI feature which might come in very useful if you do want to change the voices but at the same time preserve the original notebook LM dialogue as I know a lot of people do because the notebook LM voices are very widely used So if you use them too then your podcast won't sound unique So the feature I want to show you is the upload script option which allows you to swap out the voices for any of the voices in the Jog AI voice library And you can even edit the script to make them say whatever you want So let's say the original dialogue has a section you don't think is interesting Well you can just remove it Or let's say you think it's missing a vital piece of information Well you can just add that in So now I'll show you that process step by step Step one is to generate a notebook LM audio conversation No need to repeat this step The audio we already generated will do fine Step two is to go to jog AI Click on the video podcast section again and this time choose upload script instead of upload audio We select our audio source and wait for jog to process the audio and extract the transcript Step three is to revise the transcript check for errors and make any modifications you desire I'll just modify this part here for purposes of demonstration Step four is to choose or create an avatar pair and select the voices you want to use Adventure awaits those who dare to follow their dreams Welcome to a smarter tomorrow powered by innovation Then the final step is just to hit the render button So the final video is generated And here we have the final output and we can watch it back You know that feeling when a new wave of technology hits and you just sense things are about to fundamentally change Oh yeah sure What did you have in mind We're diving head first into one of those waves today Something people are calling vibe coding Exactly Yeah We've been looking at some really fascinating sources describing it Vibe coding It's uh it seems to be a real paradigm shift in how software gets made Quick heads up I've added a few free resources in the description You'll find a complete guide to this video a complete chat GPT course from beginner to advanced And last but not least an AI tools guide if you're still figuring out the difference between chat GPT and Copilot for example I call this my AI tools ecosystem All of the above is free So we've seen how to generate a video from the notebook LM audio both keeping the same voices and changing the voices However I want to mention that Jog AI allows you to make a video out of any podcast audio not just a notebook LM audio So for example if you publish podcasts whether real or AI by using Jog AI you can generate a video to accompany your podcast episode to publish on video platforms to expand your reach I'll show you what that looks like I have here a podcast episode I made on Jellod my favorite AI podcast generator If you don't know Jellod then see the video that appears on the card Now Jellod is an all-in-one AI podcasting platform that takes care of your script your virtual hosts generating the audio publishing and distribution all in minutes All you have to do is feed it your source information just as you would with Notebook LM It's well worth checking out and you'll find a link in the description below which will get you 15% off any Jell Pod plan for the first year Anyway let's have a little listen to the episode Hello everyone Welcome back to another episode of Cosmological Conundrums The concept of a warp drive is one of those ideas that blurs the lines between science fiction and theoretical physics Essentially the ship doesn't move through space in the traditional sense The space around it moves instead Right It's like surfing You're not the one moving through the water You're riding the wave that's pushing you forward But instead of water it's spacetime doing all the work Precisely Okay So now to generate a video from this podcast step one is to generate our podcast episode I already have mine prepared but you generate it however you would normally do it Step two is to go to Jog AI choose video podcast and then choose upload audio Select your audio file and click the next button Step three is to choose your avatar pair just as we did with the notebook LM audio In this case as the male starts speaking first I can choose any of the pre-existing pairs So I'm going to choose Trevor and Catherine here We don't need to select voices because we're generating the video direct from the existing audio And anyway I created my hosts to my custom specifications within Jellod So then all that's left to do is to hit render and wait for our result And here it is So let's take a look Hello everyone Welcome back to another episode of Cosmological Conundrums The concept of a warp drive is one of those ideas that blurs the lines between science fiction and theoretical physics Essentially the ship doesn't move through space The space around it moves instead Right It's like surfing You're not the one moving through the water You're riding the wave that's pushing you forward But instead of water it's spacetime doing all the work Precisely And this idea got its first serious theoretical foundation in 1994 with the Alcubier drive by creating what we call a warp bubble Inside this bubble the laws of physics as we currently understand them remain intact Okay so the spacecraft stays in its own little safe zone Kind of like a like a force field bubble That's pretty cool But there's a catch right Well yes The theory What did you think of that generation Let me know in the comments below I'd love to hear your opinion The last thing I want to show you is how Joerger AI can generate talking avatar videos from pretty much any source If we come back to the video podcast section again then you see the generate script option is the option that's selected by default And below the generate script option are four different source types i.e website YouTube PDF and long text So that means you can provide any web page URL or YouTube video link and it will get processed by Jog AI You can also provide a PDF or just paste in simple text It's important to bear in mind though that this feature doesn't produce a word forword video from your source What it does is create a video based on the source but with an original dialogue Also important to note is that you have to set the duration of your generated video and the max length of the generated video is 5 to 10 minutes So if your source is longer than that it will be summarized So the process is as follows Step one prepare a source I'm going to prepare this video of mine about Jellod That's where Jelly Pod steps in Step two go to video podcast in Jog AI Choose generate script Choose YouTube as the source type and paste the link Step three choose language video duration and turn duration This is how long each speaker should talk for Step four click the next button and wait for the script to be generated Step five as before choose an avatar host pair and select your voices Step six send the video to be rendered and sit back while the magic happens So let's take a look at what Jog AI has given us So have you seen all those podcasts made with Notebook LM lately They're freakishly realistic People are literally uploading AI convos and calling it a show Yeah it's kind of wild But have you tried customizing one Total nightmare You can't change much Scripts fixed voices are limited and forget about making it sound like you That's where Jelly Pod flips the script right Full creative control from content to host voices You can even clone your own voice Exactly Imagine a podcast where the host sounds like you thinks like you and delivers exactly what you want to say That's Jelly Pod And there you go We have our video podcast ready to go And this process is the same for all other source types And that's your crash course on how to automatically generate talking avatar videos for your notebook LM or other podcasts If you don't have time to manually produce talking avatar videos from your audio podcasts then Jog AI could really help you out because in just a few clicks you have your talking avatar video podcast ready to publish Just a note on rendering time Jog AI gives paid users about half an hour of priority rendering time So that first half an hour is very fast After that it will vary Now just final thoughts about the overall quality The first video we did which was directly from Notebook LM is the one that had most of the issues The other three because they didn't come directly from a notebook LM audio were smoother And I think the reason for that is that in the notebook LM audio there are a lot of interjections So that's when the other speaker says "Mhm." And it's it's a very short change from one speaker to another and perhaps even overlaps And that seemed to be the issue the thing that was causing the the first video that I showed you to have those strange effects on the cuts It didn't seem to be able to handle that quick cutting Whereas the other three videos were a lot smoother That may have happened occasionally but not as much In any case uh this is something that is improving rapidly So watch this space Now Jog AI doesn't only do video podcasts It has a whole array of features So I invite you to check them out But did you notice that in the video podcast section there's an option to add yourself as one of the avatars So that could be a very interesting feature and I know a lot of people are asking about that I'm actually doing a video on that So if you want to see the results of that video and how that turned out then we'll be coming up on screen now when it's ready Thanks so much for watching If you like this please consider subscribing I'll see you on the next one [Music]