Your whole back catalogue, transcribed.

Assuming 45 minutes an episode, 200 episodes is 9,000 minutes. Two $24 packs buy 10,000 minutes for $48 and leave 1,000 spare; one $24 pack plus two $12 packs is exactly 9,000 for the same $48. Substitute your own average episode length and the three packs do the arithmetic: $8 buys 1,000 minutes, $12 buys 2,000, $24 buys 5,000.

At 45 minutes an episode, 80 episodes is 3,600 minutes, which one $24 pack covers on its own — $24, with 1,400 minutes spare. Nothing cheaper reaches 3,600: $12 + $8 + $8 gets you 4,000 minutes but costs $28. 400 episodes is 18,000 minutes: three $24 packs, a $12 and an $8 make 18,000 exactly for $92, where four $24 packs would be $96. Credits never expire, so spare minutes go to next season.

No. There is no feed crawler that walks your show and ingests it. You paste each episode URL, import a Google Drive folder so every file in it goes in at once, or submit the list through the API. The Drive folder import is usually the fastest route for an archive that already has its masters in one place.

Put the files in one Google Drive folder and import the folder — every file in it is submitted in a single action. If your episode list lives in a CMS or a spreadsheet, take an API key and submit them programmatically instead. Both routes land in the same Library as a paste would.

Processing usually runs at about 2 minutes of wall clock per hour of audio, so a 45-minute episode is normally back in around a minute and a half. The 150 hours in a 200-episode archive is roughly five hours of machine time at that rate. You do not sit and watch it — episodes appear in the Library as they finish.

No. Transcripts are not separated by speaker, so a two-person interview comes back as one punctuated, timestamped body of text rather than labelled turns. If your workflow depends on knowing who said each line, this will not do that job. Everything else on this page — the text, the timestamps, the summary, quotes and chapters — applies to interview episodes exactly as it does to solo ones.

Yes, and you should. Your first transcript is free at any length, with no card and no account needed to start, and the second is $0.99 once per email address. Run your longest, messiest episode first — that is the one that tells you whether the output is worth $48 to you.

One file can be up to 5 GB and up to 12 hours. Anything longer is split before processing, and you do not do the splitting. Language is detected automatically across 99 languages, so a show that switches languages between episodes needs no setting changed.

It can be. Workspace is $19 a month and includes unlimited transcription in the web app, so an archive you work through inside one billing month may cost less that way than the $48 to $92 of packs. Packs are the one-time route, and the one that suits submitting through the API or spreading the work over a year.

No. Transcripts are never used to train AI models. Your archive sits in your Library, exportable as TXT, SRT, VTT, DOCX or JSON whenever you want it out.