Add AI Avatars to NotebookLM Video Podcast

0:01 Today's question is how can I create a video podcast with AI that looks super realistic. In this video I will show you exactly what you need to do, to create your own video podcast in five simple steps. But before we dive in let's first take a look at the end result. And the best part is, a lot of it boils down to simple cyber hygiene. So like basic stuff that everybody can do. Right like washing your hands for your Digital Life. Precisely. The report lays out five key steps and one of the most important is multi-factor authentication or MFA. Right. Think of it as like a double lock on your digital doors. Okay. So instead of just having a password Right. You're also going to have like some other thing instead of just a password you need an extra layer of verification. Like a code sent to your phone or something. Exactly. The report actually

0:46 shows that implementing MFA can reduce the risk of compromise by a whopping 99.2%. Yeah that's huge! That's amazing. So MFA is a no-brainer. Yeah for sure. I think this looks pretty realistic to be honest. Maybe the male Avatar looks a little more human than the female though. But let's dive in and let me show you guys how to create a video podcast like this. So the first step is of course to decide on the topic for your podcast. You can basically pick any topic you want as long as you have good content, that you can use as Source. Content that you've created yourself or content that's already out there on the internet maybe. And with the tool I use you can use documents

1:32 that you have on your Google Drive, links to websites, text documents or even a link to a YouTube video. In this case I will be using two security reports from Microsoft, and they're in PDF format. And they're actually really useful if you want to learn how to protect yourself or your business. I've put the links to both reports in the description below to create the audio file, I use Google's NotebookLM. It's a very easy to use tool and the podcast it's producing, sounds super realistic and human. So now we go to the Google Notebook LM page and you just log in with your Google account. You don't have to pay for a subscription or anything; it's just for free.

2:18 And next step is to click on, create new, and you want to upload your resources. So let's do that: the screen opens and here you see that I've have already downloaded the reports from Microsoft, so I'm selecting them both click on open. And you can see here on the left it's... yeah they're already there. And now I can do things. Now before I move on, I want to rename this notebook so I can.. let's give this a logical "security podcast" and you can see this pops this screen pops up. Here below and you can do a lot of things with the Notebook LM. For now I'm just interested in creating the in-depth discussion with Two Hosts. You can adjust a few things if you want.

3:09 So I don't know updating software and let's click on generate and it says: now this may take a few minutes, you don't have to wait so yeah. And now the recording is done so let's listen to the end result. All right. So you want to know more about MFA and uh well keeping software updated. Buckle up, because we're diving deep into some fascinating reports from Microsoft no less. Oh this is going to be good! Yeah we've got two executive summaries from Microsoft Digital Defense here one from October 2023 and a more recent one from, let me see October 2024. Okay, so sounds really good and let's move on to the next step. Now the

3:55 audio file we just downloaded contains both the male and the female voices in one track. In this next step I'm going to separate both voices and create two different audio tracks for them. You can do that with a lot of different tools of course, in my case I already use audacity, and know how it works. So that's what I'm going to use now. So I've opened the download file the audio file that the notebook LM created here in audacity. And the first thing I want to do is create a new mono track. And what we want to do is is select all the parts, that from the female, to this new track. So I'm calling this "female" and the original one I'm calling "male". So and

4:44 now it's a matter of listening to the audio file, and filtering: The best part is a lot of it boils down to simple cyber hygiene. So like basic. Here she starts talking and then I press CTRL-i to cut this up. It's like basic stuff that everybody can do. Right like. So until here CTRL-i. And then, what you want to do, is drag this over here and make sure that everything is in line. And you do that for the complete audio file. And after you've done this, this will cost a few minutes of course, you have to select all the female voices and when you're done, you will do, File, Export audio,

5:30 export to computer, and then it's important to select ' multiple files' here. The rest is all good let's see export and you see it will create a male MP3 and a female MP3. So okay. And now you've separated the voices; that's it. Now that we have separated both voices into two different audio files, it's time to use them to create the video where we see two people talking. And I want the things more realistic by using multiple camera angles. So if we want to have 2 angles for the female and 2 for the male avatar, that means that we have 4 shots in total. And with Heygen you can choose a lot of different AI Avatars. And I found 2 that offers a template

6:18 where they are in the same setting (as in the same room) but then from different angles. So here we are with Heygen. I already logged in with my account so let's hit the 'create video' button. We want to choose for an 'Avatar' video and we want to have it in 'landscape'. Okay, and the next thing you want to do is, upload your audio file that we just created so let's start with the Male. Yep, here it is and then we select a nice avatar for this, and let's see, where is the guy that I selected? Here he is: Emmanuel. And as you can see, he's sitting here on the couch. We

7:08 can select that. And he also sits here, but then from a side angle. So let's pick this one to start with. You can. "And the best part is a lot of it boils down to simple cyber hygiene" Yeah, this is the correct audio file that I just uploaded because you don't hear the female. You don't see the Avatar moving yet but that's only after you hit the submit button. So when I click that, the thing will start producing. Okay that's all there is to it. Let's move on with the showing the other example; so let's go back we hit the 'create video' button again. We choose for 'Avatar video' again, for 'landscape' and now we do the exact same thing; we upload the female voice over here.

8:02 Yep, there it is, and we pick the female Avatar. Where is she? There's a lot of avatars! ah, her name is Piper. And here you can see, you know, this is maybe not suitable, but this is a good video shot, And this is from another angle. And so, it's the same thing let's listen to if it's the right audio. "So basic stuff that everybody can do" "basic stuff that everybody can do" Yeah, this is the right one. And hit 'submit button' again and then it's going to produce. Now I fast forwarded the video a bit for you guys. But here you see the end-result of what Heygen produced. I created two videos for the female angle B and the normal angle

8:53 and I also created two videos from from the male sitting on the couch. You know, from the normal angle and with a B angle. And with these videos, I'm going to work in the next step. So now that we have 4 videos, the final step is to edit those 4 into one video. And of course I want to show the male Avatar when he is speaking, and the female avatar when she does the talking. And, as said, I want to blend in different angles to make things a little more interesting. Now DaVinci Resolve has built-in multi- camera support, so that's what I'm going to use right now. So here we are at DaVinci resolve, and the first thing you want to do is import the media that we are going to use in the this final edit. So let me drag in the four output files from Heygen,

9:46 together with the original audio file. Change project frame rate? Yeah, we should do that. And now it's importing the 4 clips that you see here and I'm also importing the original audio file, and I will explain later why I'm doing that. So now hit the 'edit' button. Yep here we are, we select the four Clips. All four, and then we hit the with our right Mouse click, and then select 'create new MultiCam clip' using the selected Clips. Okay, and let's call this 'FINAL EDIT'.

10:31 Hit 'create' and now you see it has created this new clip where all the angles are in. And you can find the original Clips in this folder here. Okay, and now we're going to drag this one into our timeline. There it is. Now, let's drag this all the way to the left, and now add the audio file as well. And you can see that here is all the audio available and here are some blank spots, because we deleted that one for the male clip. And you can normally select the the right audio, you

11:20 can see it: angle 1, angle 2, angle 3, angle 4. But now we want to delete this one. Delete track, so we only hear the ORIGINAL audio. Okay, now what we want to do is select the right angle and but we do that by clicking on multi cam. And now you see all the angles we have and now it's just a matter of playing. And here you see the output and you see when I play this. The normal angle one is selected and we want to switch to the right angle when for instance the woman is talking. Now and you do that for your entire clip; you select all the right angles and when you're done you go to

12:09 'deliver' and let's select YouTube here in this case and add it to the render queue. Okay, let's call this 'final edit' save this one. And then you can start the render process by clicking on render all. And that's all there is to it! What else can businesses do to kind of bolster their defenses? Well another critical step is keeping your systems updated. Okay. You know think of it like patching cracks in your walls before the rain gets in. Hackers love to exploit vulnerabilities in outdated software.. Now I hope that this video can help

12:59 if you want to create your own video podcast with AI, and if you have other requests for videos? Let me know in the comments, and do not forget to subscribe! Thank you....