Start transcribing
Sign up now to start transcribing your audio and video files with high accuracy, secure and private processing, and industry-leading features like speaker diarization and subtitle generation.
Upload an interview – audio or video – and get it back as a dialogue with the speakers marked, ready to quote or code. Interviews up to 10 minutes are free without an account; speaker labels and full-length interviews come with a free account (30 free minutes), then you pay per minute – no subscription. A one-hour interview costs about $2.95 on the smallest pack.
Speakers marked: each change of speaker is labelled [SPEAKER_00], [SPEAKER_01] …, so an hour of conversation reads as a dialogue, not a wall of text.
Quote-ready: edit the text online, download TXT, or SRT/VTT with timestamps to find a passage in the recording.
Pay per hour, not per month: one interview or fifty – buy the minutes you need; they never expire.
Language detected automatically: interviews in Spanish, French, Arabic or German need no setting; 55 languages detected on this site so far.
Confidential: files are processed automatically, never shared, deleted after 30 days; delete them earlier yourself.
Free up to 10 minutes without an account · 30 free minutes with one.
Supported formats: MP3, M4A, WAV, MP4, MOV, OGG, FLAC · Prices
Interviews are what people pay us for. Among 10–30-minute recordings from registered users, 30 % have exactly two speakers and 29 % three or four; a two-person interview averages 13 minutes and comes back split by speaker. Researchers at NYU, Berkeley and 400+ other universities use it; journalists at Hearst and 80 media organisations.1
1 People at these organisations use transcribe.mov with individual accounts; this is not an institutional endorsement.
Facts updated August 2026
On OpenAI’s published benchmark of read speech (Common Voice 15) the large-v3 model makes about one error in 11 words in English, one in 21 in Spanish, one in 18 in German, one in 17 in Portuguese and Russian, one in 9 in French. On clean studio recordings (FLEURS) it is roughly a third better. Independent testing on real meetings and calls (Open ASR Leaderboard, 2026) puts English at 7–11 % word error. Long recordings are conversations, not read speech: on the independent long-form test the figure is ≈ 11 % word error. transcribe.mov runs the faster “turbo” build of the same model, which is a few tenths of a point behind on English. Real files are harder than benchmarks: background noise, cross-talk and strong accents add errors, and a wrong language guess ruins a transcript.
Sources: OpenAI large-v3 release chart (Common Voice 15, FLEURS), Open ASR Leaderboard paper 2026-03-30.
After the speech is transcribed, a second model listens for changes of voice and assigns each segment to a speaker. The labels come back as [SPEAKER_00], [SPEAKER_01] and so on, in the order the voices first appear. Two or three clearly different voices in a quiet room are separated reliably; similar voices, crosstalk and a speakerphone at the far end of a table produce more mistakes.
In the online transcript, behind the “Show Speakers” switch, and in the JSON export, where every segment carries its speaker. The TXT download and the SRT/VTT subtitle files contain the text and the timing without speaker names – if you need labelled text in a document, copy it from the online transcript.
One microphone between the two of you on a quiet table is enough; a phone works. Avoid speakerphones and cafés, and ask the interviewee to let you finish a question before answering – overlapping speech is the main source of merged turns. A 60-minute interview recorded this way needs a quick read-through, not a re-listen.
Five lines from a transcribed interview as the online transcript shows them with “Show Timecodes” and “Show Speakers” switched on. This is the raw output before any editing.
[00:00:00 --> 00:00:05] [SPEAKER_00] Thanks for making the time. Can you start with how the team is set up? [00:00:05 --> 00:00:11] [SPEAKER_01] Sure. There are six of us, and three of them joined this year. [00:00:11 --> 00:00:14] [SPEAKER_00] And what changed when they came in? [00:00:14 --> 00:00:19] [SPEAKER_01] Mostly the handover. We used to lose a day every week to it. [00:00:19 --> 00:00:22] [SPEAKER_00] A day a week. What did you do about it?
The JSON export carries the same labels with the timing of every word; TXT, SRT and VTT contain the text without labels.
The transcript marks each change of speaker and labels the turns [SPEAKER_00], [SPEAKER_01] and so on, in the order the voices first appear. Speaker labels are included for registered users; guests get plain text.
For a clear recording it produces a near-verbatim transcript that needs a read-through rather than a re-listen. Check quotes against the recording before publishing – the timestamps in the SRT/VTT export take you straight to the passage. Filler words are mostly dropped, so it is not a strict verbatim transcript.
Copy the text from the online transcript with “Show Speakers” switched on and paste it into your document, or use the JSON export, which contains the label of every segment. The TXT download itself has no labels.
Yes – 30 % off every pack for non-profits, educators, researchers and students. Email contact@transcribe.mov from your organisation’s address and you will have your discount code within a day.
Up to 10 minutes and 50 MB without an account, free. A free account gives you 30 free minutes and files up to 2 hours and 4 GB. Beyond that you buy minutes: 120 for $5.90, 300 for $10.90, 600 for $19.90 – a one-hour interview is about $2.95 on the smallest pack, one-time, no subscription.
Yes. The language is detected automatically – 55 languages so far on this site, with Spanish, French, Portuguese, Tagalog, German and Arabic the most common after English. An interview that switches between two languages is transcribed in the one that dominates.
Nobody but you. Files are processed automatically, never shared, deleted from our servers after 30 days, and you can delete a transcript and its file at any time from your account.
Stripe emails a receipt for every purchase; if you need an invoice for your records, email contact@transcribe.mov and you will get one.
transcribe.mov uses Whisper large-v3 (the turbo build) – the open model that OpenAI benchmarks per language. On read speech it makes roughly one error in 11 words in English and one in 18–21 in Spanish, German, Portuguese or Russian; clean studio audio is better, noisy or overlapping speech is worse, and independent tests of real meetings land around 11 % word error. You get the whole file, verbatim, with punctuation and speaker labels (registered users); you fix names and terms in the editor.
Record close to the speaker in a quiet room – one voice at a time transcribes best. Upload the original file; don’t re-encode or compress it (phone .m4a and camera .mov are fine as they are). Clear speech matters more than file quality: accents and cross-talk cost more accuracy than bitrate ever will. Silence and music are filtered automatically.
For a short clip you can. Gemini takes audio and video uploads, but its free tier stops at 10 minutes of audio or 5 minutes of video per prompt and needs a Google sign-in (3 hours of audio with a paid AI Pro plan). ChatGPT has no documented audio or video transcription feature: it may try on an uploaded file, and OpenAI itself says the analysis "may not analyze the entire video or accurately interpret its audio". In both, timestamps are written by the language model, not measured, so they drift, and there are no speaker labels on uploaded files. Here the interview is transcribed in full and verbatim by a speech model, the SRT/VTT timing comes from the audio itself, speakers are labelled, files up to 10 minutes need no account at all, and your files are never used to train anything.
Facts updated August 2026
Sign up now to start transcribing your audio and video files with high accuracy, secure and private processing, and industry-leading features like speaker diarization and subtitle generation.
transcribe.mov - Free, Accurate, and Instant Transcription for Your Audios and Videos! Unlock the Power of Our AI-Powered Transcription Tool.
