Podcast transcript generator
Most podcasts never publish a transcript, which makes a great episode oddly hard to quote, search, or skim. Paste an episode page, an RSS feed, or a direct audio link below and tautau transcribes it with the same speech models that power its dictation. A free account includes 15 minutes of audio every month, no credit card, and paid plans raise that to 200 or 600 minutes.
YouTube videos with captions come back instantly: 5 free transcripts with no account, 5 a day with a free account. Podcast episodes and direct audio files are transcribed by AI: free accounts get 15 minutes a month.
How it works
- Paste the link. An episode page, the show's RSS or Atom feed, or a direct link to an mp3, wav, m4a, webm, or ogg file.
- The AI transcribes it. The audio runs through a hosted speech model. This is real work, so a long episode takes a few minutes while the page polls and shows progress.
- Use the transcript. Search, edit, and copy it in the workbench, or download it as .txt or .md.
What a transcript unlocks
Quoting a guest accurately instead of transcribing by ear. Pulling show notes out of a rambling two-hour conversation. Searching across a back catalogue for the episode where they mentioned your topic. Reading an episode at your own pace when listening is not an option. The Analytics tab in the workbench adds word frequency and stats on top, and signed-in transcripts land in your private history so they are there when you come back. For YouTube videos there is a faster path: the YouTube transcript generator reads the caption track directly and is free to try without an account.
Honest expectations
AI transcription is good, not perfect. Clear studio recordings come out clean; a noisy remote interview with crosstalk will need a proofread, and proper nouns are the classic weak spot of every speech model. Long episodes also take real time, measured in minutes rather than seconds. The free tier's 15 monthly minutes cover the occasional episode; if you transcribe every week, the paid plans are priced for that. More detail on formats, feeds, and quotas lives in the docs, and the main transcript tool handles every source type in one place.
Frequently asked questions
How many free minutes do I get?
A free account includes 15 minutes of audio transcription every month, no credit card required. Paid plans raise that to 200 minutes (Base, $3/mo) or 600 minutes (Pro, $9/mo), shared with tautau dictation. Anonymous visitors are asked to sign up first, because speech models cost real compute.
Can I paste an RSS feed, or does it have to be an episode link?
Either works. Paste the episode page URL and tautau finds the audio behind it, or paste the show’s RSS or Atom feed and the latest episode is picked up. Direct audio file links work too: mp3, wav, m4a, webm, and ogg.
How long does a transcript take?
Longer than a YouTube caption fetch, because the audio is genuinely transcribed by a speech model rather than read from an existing track. Short clips come back quickly; a full-length episode takes a few minutes. The page shows a progress state and checks for the result every few seconds, so you can leave the tab open and come back to a finished transcript.
Which audio formats are supported?
mp3, wav, m4a, webm, and ogg, plus whatever enclosure format a podcast feed points at. Most podcast hosting serves mp3, which is fully covered.
Where does my transcript go?
Nowhere you do not put it. The result opens in the workbench, where you can search, edit, copy, or download it as .txt or .md. If you are signed in, the transcript is also saved to your private, searchable history and you can delete it at any time. Audio is used only to produce the transcript.
Or skip the keyboard entirely
tautau puts a mic button next to every text field on the web. Click it, speak, and your words land in the box with punctuation handled. A free account gives you 5 dictations a day, no credit card required.