VIDEO
HIGHLIGHT
Log InSign up
Turn NotebookLM into Videos with YOUR Face (NO FILMING!)
SummaryTranscriptChat

Turn NotebookLM into Videos with YOUR Face (NO FILMING!)

Transforming Notebook LM Podcasts into Engaging Video Content

Introduction to Notebook LM and Its Limitations

  • Notebook LM can generate realistic AI conversations, but lacks video content, limiting distribution and engagement.
  • The discussion introduces a method to convert Notebook LM audio into hyperrealistic talking head videos using a specific tool.

The Importance of Video in Podcasting

  • Without video, creators miss out on major platforms like YouTube and TikTok, leaving potential audience engagement untapped.
  • The presenter emphasizes the effectiveness of the Design tool for creating synchronized videos with natural movement and believability.

Step-by-Step Process Overview

  • Users can utilize their own face or an AI-generated one for the podcast video; permission is required for real faces.
  • The first step involves generating an audio dialogue using NotebookLM by uploading source content.

Generating Audio with NotebookLM

  • NotebookLM analyzes various sources to create multimedia reports, including natural-sounding AI podcasts.
  • Users can set parameters for podcast duration before generating audio content.

Preparing Visual Elements for Video Creation

  • A crucial step is preparing images of podcast hosts in a shared scene to ensure quality in the final video output.
  • Using AI image generators like ChatGPT or Nano Banana helps create high-quality host images that maintain facial likeness.

Utilizing Design Tool for Video Production

  • After signing up on Design, users upload prepared host images and select them for animation within the platform.
  • It’s essential to split audio files into separate tracks for each host to avoid simultaneous speaking issues during playback.

Finalizing Video Content

  • Speaker Split allows users to easily separate multi-person dialogues into distinct audio files necessary for synchronization in Design.
  • Once speech tracks are loaded correctly onto their respective timelines, users can preview before generating the final video product.

Reviewing Generated Videos

  • The generated videos exhibit impressive realism with hosts displaying natural movements and expressions throughout discussions.

Enhancing Voice Quality with Cloning Technology

  • To match visuals with voiceovers effectively, users can clone their voices directly within Design without needing external tools.

Conclusion: Future of Podcasting with AI Integration

  • This process enables creators to launch professional-looking video podcasts effortlessly while maintaining brand consistency across episodes.

The Art of Natural Dialogue in Podcasting

Creating Human-Like Conversations

  • Real conversations often include interruptions and overlaps, which make dialogue feel authentic. Descript's timeline feature allows users to replicate this natural flow by placing speech segments freely.
  • Human behavior is influenced by hidden patterns such as fear, status, belonging, and social imitation, which can lead people to believe in seemingly absurd ideas.

Balancing Chaos with Structure

  • The speaker humorously suggests that while chaos can be entertaining (e.g., "chaos snacks"), it’s essential to provide structure and evidence during discussions for clarity.
  • Overlapping dialogue is encouraged; hosts can interrupt each other or respond spontaneously without limitations, enhancing the realism of the conversation.

Understanding Video Generation Costs

Credit System Breakdown

  • A 5-minute video generation costs approximately 2,000 credits at 720p or 2,600 at 1080p. This serves as a benchmark for evaluating different plans.
  • Users on the Master plan receive four videos per month at 720p or three at 1080p. The Master Pro plan offers more flexibility with up to 14 videos at lower resolution.

Maximizing Credits Usage

  • For short clips (10 to 20 seconds), the Creator plan provides up to 25 clips for 3,000 credits. As AI generation costs decrease, longer video durations may become feasible in the future.

Tools for Enhancing Podcast Production

Essential Features for Podcasters

  • Note Book AI assists with audio management while Speaker Split separates speakers effectively. Design helps create visually engaging content using your face and voice.
  • Viewers are encouraged to subscribe and engage with feedback or questions regarding the tools discussed. Complete show notes are available for further reference.
Video description

Try dzine free: https://www.dzine.ai/?via=jonathansjam Turn your NotebookLM podcast into a talking-head video — with your face, your voice, and realistic AI hosts that are genuinely hard to tell apart from the real thing. NotebookLM can generate hyper-realistic AI podcast audio every single day on any topic. But without video, you're locked out of YouTube, TikTok, and Reels — and leaving most of your potential audience on the table. This video fixes that in three steps. I've tested every AI podcast-to-video tool on the market. For creating a NotebookLM talking-head video with your own face on it, nothing comes close to Dzine — the lip-sync quality, likeness, natural movement, and overall believability are in a league of their own. What you'll learn: How to generate your AI podcast audio in NotebookLM How to split multi-speaker audio by host using SpeakerSplit How to create a NotebookLM talking-head video in Dzine with your face and cloned voice How to animate AI-generated characters if you don't want to use a real face Dzine's hidden superpower for host believability that no other tool has ============= *LINKS MENTIONED IN THE VIDEO:* *Dzine* Get started free and put your own face on a podcast: https://www.dzine.ai/?via=jonathansjam *NotebookLM* Understand anything: https://notebooklm.google/ *SpeakerSplit* for automatic separation of speakers: https://speakersplit.io/?via=jonathansjam Get 10% off with my discount code. *HUBJAM10* 📝 *Show Notes* – all links + tools from the video https://jonathansjam.com/go/show-notes/ *Full-length generations done with Dzine* Here are my complete generations for you to check-out (let me know what you think). https://youtu.be/1d34hMONd4M ============= *📖 VIDEO CHAPTERS:* 00:00 Introduction 02:19 Method 1: Turn NotebookLM podcast into a video (keeping original NotebookLM voices). 02:20 Step 1: Create NotebookLM podcast 04:51 Step 2: Create image of podcast hosts 07:20 Step 3: Split the speakers into separate files 09:08 Step 4: Create talking head video with your own face 11:34 Finished 5min render with Jonathan and Silvia 13:23 Excellent head and hand gesture examples 14:26 Method 2: Put your own face and voice on a NotebookLM podcast 15:08 Step 1: Clone your voice 16:05 Step 2: Transcribe the speech into text 17:45 Step 3: Generate speech with your new voice clone 20:30 Finished render with Jonathan (+ voice) and Silvia 21:28 Alternative generations with my face 22:49 Generating hosts image inside dzine (non-human hosts) 24:33 Overlapping speech in dzine 26:35 dzine pricing ============= *Sponsorship disclosure* Thanks to dzine for sponsoring this video. *Affiliate disclosure:* Some links may earn me a small commission at no extra cost to you. I only share tools I trust and use. *All thoughts are my own.* Thanks for supporting the channel! Here are your hashtags: #NotebookLM #NotebookLMVideo #NotebookLMTutorial #NotebookLMPodcast #AIVideo #AITalkingHead #TalkingHeadVideo #AIPodcast #AIPodcastToVideo #PodcastToVideo #AIVideoGenerator #VoiceCloning #AIVoiceCloning #LipSync #AILipSync #Dzine #DzineAI #DzineTutorial #AIContentCreation #AIContent #AITools #AITools2026 #AIVideo2026 #ContentCreation #ContentCreationTips #YouTubeGrowth #AIForCreators #FacelessVideo #AIFaces #RealisticAI