Paste an episode link or upload the audio, and get a timestamped transcript you can read, search, quote and export. Works across Apple Podcasts, Spotify, YouTube, SoundCloud and direct MP3 URLs.
Open JotVox →It turns an episode of spoken audio into text with timestamps, so you can read it, search it and copy exact lines instead of scrubbing back and forth through a player.
JotVox does this two ways. If the episode already has a published caption or transcript track, it grabs that text. That route is free and needs no sign-up. If there is no existing transcript, which is still the normal case for podcasts, JotVox transcribes the audio with AI at 98%+ accuracy across 99 languages, using credits.
Every line carries a timestamp. That is the part that matters for podcast work: you can cite minute 42 rather than telling someone the quote is somewhere in the back half.
It runs in the browser. There is no extension to install and nothing to download.
JotVox takes a link or a file, not one specific app. In practice that covers nearly every way a podcast gets published:
If a show sits behind a paid member feed, download your own copy and upload the file instead. For the platform-specific routes, see Apple Podcasts transcripts and SoundCloud transcripts. The full list is on supported platforms.
Plan for real time on long audio. A ninety-minute interview is a lot of speech, and AI transcription of an hour-plus episode is not instant. Start it and come back rather than watching the screen.
On cost, one credit is one action. Signing up gives you 3 free credits, and once you verify your email you also get one free AI transcription a day, which is enough to work through a back catalogue slowly. Credit packs start at $4.99 for 20 credits and subscriptions start at $9 a month. Grabbing a transcript that already exists costs nothing.
Audio you upload is processed in memory and deleted straight after transcription. That is worth knowing if you are working on an unreleased episode, a client's recording or an interview you have not cleared yet.
More on pacing long files: transcribing long videos and podcasts.
Clean solo narration transcribes close to perfectly. Conversation is harder, and it is worth knowing where errors land before you paste a quote into something you publish.
Background on what drives the numbers: AI transcription accuracy.
Five jobs come up again and again:
Exports are TXT, SRT, VTT and MP3, so the transcript travels into whatever tool you write in.
1. Get the episode link or file Copy the episode URL from Apple Podcasts, Spotify, YouTube or SoundCloud, or the direct MP3 address out of the show's RSS feed. For a members-only episode, download the audio and have that file ready instead.
2. Paste it into JotVox Drop the link in, or upload the file from your device, Google Drive or Dropbox. If the episode has a published transcript track, JotVox grabs that text free with no sign-up. If it has none, AI transcription runs on the audio using one credit.
3. Let a long episode finish An hour-plus episode takes real time to process. Leave it running and come back instead of reloading the page.
4. Clean it up and export Cut the sponsor read, check every guest name and company name against the episode page, then export to TXT for writing or SRT and VTT for captions.
Yes. Paste the Spotify episode link into JotVox and it returns a timestamped transcript. If that episode has a published transcript track, the text is grabbed free with no sign-up. If it does not, JotVox transcribes the audio with AI at 98%+ accuracy using one credit. The same works for Apple Podcasts, YouTube and SoundCloud episode links, or for an audio file you upload.
Long audio takes real processing time, so treat an hour-plus episode as something you start and come back to rather than wait on. Grabbing an existing caption track is quick, because the text already exists. Full AI transcription has to work through every minute of speech, and a ninety-minute interview is a lot of minutes. Start it, then go do something else.
Grabbing a transcript that already exists is free and needs no sign-up. AI transcription, for episodes with no published transcript, uses credits. You get 3 free credits when you sign up, plus one free AI transcription per day once your email is verified. Beyond that, credit packs start at $4.99 for 20 credits and subscriptions start at $9 per month. One credit equals one action.
Often, but names are the most common error in podcast transcripts and you should always check them. Proper nouns of every kind, guest names, company names, book titles and technical terms, are where AI transcription slips, because the model is working from sound alone. Before you publish a quote or a set of show notes, verify spellings against the episode page or the guest's own profile.
Only as a file afterwards. JotVox does not join live calls and is not a meeting notetaker. If you have a saved recording as MP4, MP3, M4A or WAV, upload it and JotVox transcribes it like any other audio. It also does not take an existing SRT as input, because it is a transcription tool rather than a subtitle file converter.