DocQuest
Document insight, AI tutor, pdf, audio & video to podcasts
9 followers
Document insight, AI tutor, pdf, audio & video to podcasts
9 followers
DocQuest transforms PDFs, audio & video into intelligent podcasts. Unlike fragmented tools, we unify all formats in one platform. ✨ Multi-doc analysis (10 files at once) ✨ PDF, audio, video → Podcast ✨ 24/7 AI Agents for automation ✨ Adaptive AI Tutor for learning
The audio/video to podcast pipeline is a really compelling use case. I'm building in a similar space (turning creative input into different media formats with AI) and the multi-format ingestion is the part that usually trips people up. How do you handle long-form video content — does it chunk it into segments or try to process everything at once? And does the podcast output include generated narration or is it more of a restructured summary?
@mattias_s Great question! For long-form video, we extract the audio track first, then pass it
to our model for transcription. we splits large audio into 10 mins and chunks with 5s overlap, then transcribes each at full fidelity, and concatenates to avoid lossy compression
On the output side it's fully generated narration, not a summary.
The multi-format ingestion piece you mentioned is real we support direct file upload (audio/video), remote URLs, YouTube links and text/document collections. All feeding into the same generation pipeline.
Curious what format outputs you're targeting in your project?
@maximilien_kpizingui Thanks for the detailed breakdown! The 10-min chunks with 5s overlap is smart — avoids the boundary artifacts you'd get from hard cuts. Good to know the output is full narration rather than a summary, that's much more useful.
For my project (MusicOrb) the output is MP4 music videos — we chain Claude for lyrics/storyboard, Suno for audio generation, Gemini for scene images, and Remotion for video compositing. Different domain but similar chunking challenges when syncing lyrics to audio timestamps. Would love to see DocQuest handle music/audio analysis too someday!
@mattias_s Thanks! MusicOrb sounds awesome that pipeline is next-level
MusicOrb and DocQuest solve parallel problems in different domains. 🤝
Funny you mention music analysis we've had early users asking if DocQuest can podcastify song lyrics + liner notes for deeper artist context. Not on the roadmap yet, but your comment just moved it up the list 😉
if you ever want to turn your MusicOrb docs into a podcast for users, you know where to find me.
Keep building! 🚀
@maximilien_kpizingui Really cool that you're seeing the same chunking challenges with audio — we hit that exact wall syncing lyrics to music timestamps. The idea of podcastifying liner notes and lyrics is interesting, could add a whole new layer for artist storytelling. Would definitely be open to exploring a DocQuest integration down the line. Keep shipping!
Turning a dense PDF into something you can listen to while commuting is a genuinely useful idea.
@utkarsh_singh64 ,yes indeed thank you for your upvote