PodPast.ai

Stop forgetting podcasts. Start querying them.

PodPast.ai turns your YouTube channels and podcast RSS feeds into a searchable knowledge vault. Add any feed, get every word transcribed and semantically indexed, then query across your entire library inside Claude or ChatGPT — with citations and timestamps pointing back to the exact moment.

11.2 hrs/week average podcast consumption<20% retention without searchPowered by Claude

How it works

  1. Add a feed

    Paste any YouTube URL or podcast RSS feed. PodPast fetches the full episode catalogue — every video or episode, going back as far as the feed allows.

  2. Transcribe and index

    Every word is transcribed — free via YouTube captions or RSS transcripts; Deepgram nova-2 for audio-only episodes — and stored as a semantic vector index. Each chunk retains its timestamp and source metadata.

  3. Search anywhere

    Ask questions in Claude via MCP or use the web search interface. Every answer cites the source episode with a timestamp link — click to jump to the exact moment in the original audio or video.

Why knowledge workers switch from Snipd and NotebookLM

Both tools are per-session or per-episode. PodPast.ai is a persistent vault — it grows with your library.

FeaturePodPast.aiSnipdNotebookLMPodwise
Cross-corpus search✓ Unlimited✗ Per-episode✗ Per-notebook✗ Per-episode
Library sizeUp to 25 podcastsUnlimited50 sources/notebookUnlimited
Web access✗ Mobile only
Claude MCP✓ Native✓ Ships
Export to Obsidian/ReadwiseComing soon

PodPast.ai is the only tool in this category that supports cross-corpus semantic search across your whole library, natively integrated into Claude via MCP.

11.2 hrs

avg weekly podcast consumption
(Edison Research, 2024)

<20%

information retention
without retrieval tools

~504M

podcast listeners
worldwide

Used by knowledge workers at [companies placeholder]

Frequently asked questions

What is PodPast.ai?
PodPast.ai is a personal knowledge vault that ingests YouTube videos and podcast RSS feeds, transcribes every episode, and exposes the full corpus to Claude and ChatGPT via semantic search and a Model Context Protocol (MCP) server — so you can query across everything you've ever listened to, with timestamped citations back to the source.
How is PodPast.ai different from Snipd?
Snipd is a mobile listening app focused on per-episode highlights. PodPast.ai is a web-first knowledge system focused on cross-corpus search — you can ask a single question and get answers cited from 50 different episodes across 10 different podcasts, something Snipd's architecture doesn't support.
How is PodPast.ai different from NotebookLM?
NotebookLM caps you at 50 sources per notebook and has no persistent library. PodPast.ai maintains a persistent vault that grows with your listening — up to 25 podcasts with their full back-catalogues on Pro — and integrates directly into Claude via MCP rather than requiring you to switch to a separate app.
What transcription model does PodPast.ai use?
PodPast.ai uses a priority stack: YouTube's built-in captions and RSS transcript tags (both free), then Deepgram nova-2 for audio-only episodes. Most podcast episodes have captions or RSS transcripts, so the majority of ingestion is free — Pro includes 100 transcription minutes a month for the rare episode that needs machine transcription.
How do I connect PodPast.ai to Claude?
Install the PodPast MCP server in Claude Desktop's config file. Once connected, Claude can search your entire vault directly from any conversation — just ask it a question and it will retrieve relevant transcript chunks with timestamps. Setup takes under two minutes.
Does PodPast.ai work with ChatGPT Plus?
ChatGPT integration is in development. Currently, PodPast.ai works natively with Claude via MCP. ChatGPT support will launch as a Custom GPT — sign up to be notified when it ships.
Is there a free tier?
Yes. The free tier lets you follow 2 podcasts, indexes the last 25 episodes of each, and gives you 30 questions a month against your library — with full Claude MCP access and cited, timestamped answers. No credit card required to start.
Where is my data stored?
Transcripts and vector embeddings are stored in a Supabase PostgreSQL database hosted in the US. Audio files are never stored — PodPast.ai streams the URL to Deepgram for transcription and discards it. You can delete your vault data at any time from your account settings.

Simple, transparent pricing

Start free. Upgrade when your library grows.

Free

$0/mo

  • ✓ Follow 2 podcasts
  • ✓ Last 25 episodes per show
  • ✓ 30 questions per month
  • ✓ Full Claude MCP access
Start free

Pro

$12/mo

  • ✓ Follow 25 podcasts
  • ✓ 1,500 questions per month
  • ✓ 100 transcription minutes/mo
  • ✓ Priority transcription queue
Upgrade to Pro