Transcribe audio to text, free, in your browser
Drop an MP3, M4A or WAV recording, or another audio file your browser can play, and Burncaps writes down every word as a script. Click any word to correct it, then download the words as a text file in paragraphs, or as SRT or VTT subtitles. Text and subtitle files are free without limit for recordings up to ten minutes, with no sign-up.
The file stays in your browser. On a computer, the sound is heard on your own device unless you choose our speech service; on a phone, it goes to our speech service unless you keep it on the phone. The service gets short pieces of sound, never the file itself.
- Free, no sign-up
- MP3, M4A, WAV and more
- Text, SRT or VTT
- Click any word to fix it
Live preview, rendered on this page by the same engine that burns your video.
How to transcribe an audio file to text
Burncaps is a free audio to text converter that works inside the page. A podcast, an interview, a lecture, a meeting or a voice note becomes text in three steps, with no account.
Drop the recording
An MP3, M4A, WAV or other audio file. It opens straight on its script, with no video player to wait for, and the file stays in your browser.
Read and correct
When it has listened, every word sits on the page as a script. Click a word to correct it, and one click changes it everywhere it was heard. Type names, brands and jargon in Names to spell right and our AI spells them right in the script (only the text goes).
Download the words
Press Text for a .txt file in paragraphs, or SRT, VTT or ASS for subtitle files. Tick Times in the Text file and each paragraph starts with its time in the recording, like [00:01:23].
Where the sound goes
The file stays in your browser. On a computer, English is heard on your own device unless you choose our speech service, which is also offered for other languages and for recordings over ten minutes; on a phone the sound goes to our speech service unless you keep it on the phone. The service gets short pieces of sound and passes them to Microsoft's speech recognition in the EU, which by Microsoft's terms for the service stores none of it. Without a paid plan it hears three hours of sound in any 30 days; after that, your own device can listen instead. The script's text leaves your device only if you type names to spell right or choose a translation (it then goes to our AI); ticking Keep everything on this device stops that. Transcribe without uploading shows how to check every request yourself.
Which audio files work
Any audio file up to ten minutes that your browser can play opens. Past ten minutes, Burncaps reads the sound a piece at a time instead of holding the whole file, and that reader takes three kinds of file.
| File | Up to ten minutes | Over ten minutes, up to three hours |
|---|---|---|
| MP3 | Yes | Yes |
| M4A | Yes | Yes, with AAC sound inside |
| WAV | Yes | Yes, with plain (uncompressed) sound inside |
| AAC, OGG, FLAC or Opus | Yes, if your browser can play it | Save it as an M4A, MP3 or WAV first, or trim it to ten minutes |
One page for each common file
MP3 to text covers podcast downloads and recorder files, with what an MP3's size says about it. M4A to text covers what sits inside an M4A and the audio file Zoom saves. Voice memo to text starts with the iPhone's own transcript and covers what it does not. If you have the video of the recording too, transcribe video to text is the same job from an MP4 or MOV.
WAV to text
A WAV file holds the sound uncompressed, so it is big: an hour of CD-quality stereo WAV is about 635 MB (44,100 samples a second, two channels, 2 bytes each). That is still far under the 10 GB a long recording may be. Over ten minutes, Burncaps reads any WAV of plain, uncompressed sound, and it mixes stereo to one channel before it listens, as it does with every file.
What you get back
| File | What it holds | Use it for |
|---|---|---|
| Text (.txt) | The words in paragraphs, in the language spoken, with your corrections. A new paragraph starts where the speaker finishes a sentence and stops for about a second, or after five sentences without such a stop. With Times in the Text file ticked, each paragraph starts with its time, like [00:01:23]. | Notes, an article, a record of a meeting or an interview |
| SRT (.srt) | Numbered subtitle lines, each with its start and end time, up to two lines of 42 characters, never running across a sentence | YouTube, Vimeo, video editors and media players |
| VTT (.vtt) | The same timed lines in the web's own subtitle format | Video players on web pages |
| ASS (.ass) | The same timed lines in the SubStation Alpha format | Players and subtitle editors that read it |
What the files do not hold
No file names who is speaking, so a meeting's or an interview's transcript needs the names added by you. A translation of the lines goes into the SRT, VTT and ASS files; the Text file stays in the language spoken. The SRT file format, explained shows what each line of a subtitle file means.
How long it takes, and long recordings
Up to ten minutes, a recording's text and subtitle files are free, as often as you like. A longer one, up to three hours and 10 GB, gives its first ten minutes free (the text then ends at the last pause before ten minutes), and all of it as one of your free videos (2 every 30 days), in the 7-day Captions + Clips trial (free, no card), or with Captions + Clips or Captions. A phone takes the same recordings as a computer.
Before you drop an hour-long recording
Hearing a long recording on your own computer takes a while. On a 2020 MacBook with 8 GB of memory, 90 minutes of English were heard on the device in 8.8 minutes with WebGPU (5 October 2026); slower machines take longer. Our speech service is much quicker: before a recording over ten minutes is heard, a computer asks once which way to listen, each with its time, while your speech-service allowance covers it. Subtitles for long videos says more about long files.
Languages, and how accurate it is
It writes 59 spoken languages, the list in Listening settings. It listens first and names the language it hears above your script, with a one-click change. Twelve Indian languages, Hindi, Urdu, Bengali, Tamil, Telugu, Gujarati, Marathi, Kannada, Malayalam, Punjabi, Nepali and Odia, can be written by our speech service or on your own device by a speech model made for each (AI4Bharat IndicConformer, open source).
Accuracy, honestly
No speech model gets every word right, and how close it comes depends most on the recording: one clear voice near the microphone is written best, while music under the voice, people talking over each other, a far-off microphone or a strong accent bring more mistakes. Max accuracy in Listening settings is a larger speech model for harder audio (a 1.4 GB download, once), our speech service is the other way to listen, and every word can be corrected with a click before you download. We give no single accuracy figure for English, because it changes with every recording; for twelve Indian languages we measured both ways on public test recordings, and every number is in speech to text accuracy in 12 Indian languages.
Then make subtitles, or captions on the video
The SRT or VTT from the same script goes straight into YouTube, Vimeo or a video editor; audio to SRT is that job told in full. Captions burned into the picture and Clips work on a picture, so they need the video file: if the recording was filmed, drop the video and Burncaps makes a captioned video from its words in 60 styles, free for 2 videos every 30 days with no watermark. Or choose Clips, and Burncaps finds the best moments and cuts each into a ranked, captioned 9:16 clip; each free video exports up to 3 of them, and Captions + Clips exports every clip, unlimited. The podcast clip maker shows it on a filmed podcast.
What it costs
Free: 2 videos every 30 days, each with captions and up to 3 clips, no watermark, no sign-up. Then try Captions + Clips free for 7 days, no card. Captions + Clips is $29 a month for unlimited clips and captions; Captions is $9.98 a month or $49 once.
Questions, answered plainly
How do I transcribe audio to text for free?
Drop the audio file, such as an MP3, M4A or WAV, on this page. Burncaps hears it in your browser and writes every word as a script; correct any word with a click, then press Text for a .txt file, or SRT or VTT for subtitles. Text and subtitle files are free without limit for recordings up to ten minutes, with no sign-up.
Which audio files can I transcribe?
MP3, WAV and M4A work, and so do AAC, OGG, FLAC or Opus files your browser can play. Over ten minutes, use an M4A, MP3 or WAV file with AAC, MP3 or plain sound inside.
How long can the recording be?
Up to ten minutes, a recording's text and subtitle files are free, as often as you like. A longer one, up to three hours and 10 GB, gives its first ten minutes free, and all of it as one of your free videos (2 every 30 days), in the 7-day Captions + Clips trial (free, no card) or on a paid plan.
How long does it take to transcribe an hour of audio?
It depends on the computer. On a 2020 MacBook with 8 GB of memory, 90 minutes of English were heard on the device in 8.8 minutes with WebGPU (5 October 2026); slower machines take longer. Our speech service is much quicker: before a recording over ten minutes is heard, a computer asks once which way to listen, each with its time.
Does the text file have timestamps?
Only if you want them. The Text file is in paragraphs; tick Times in the Text file, under the file buttons, and each paragraph starts with its time in the recording, like [00:01:23]. For a time on every line, save SRT or VTT instead: each subtitle line is timed to the voice.
Is my audio file uploaded?
The file itself stays in your browser. On a computer, the sound is heard on your own device unless you choose our speech service; on a phone it goes to our speech service unless you keep it on the phone (short pieces of sound to Microsoft's speech recognition in the EU, which by Microsoft's terms for the service stores none of it).
Which languages can it transcribe?
59, the list in Listening settings. It listens first and names the language it hears above your script, with a one-click change. The Text file is in the language spoken; a translation of the lines goes into the SRT and VTT files.
Does it say who is speaking?
No. The Text file holds the words in paragraphs and does not name or label who is speaking, so for a meeting or an interview you add the names yourself.
Burncaps facts, checked 3 November 2026
What it is: a browser tool that burns animated word-by-word captions into a talking video, or finds the best moments of a talk, podcast or interview and gives them back as ranked, captioned 9:16 clips.
Price: free for 2 videos every 30 days, each one recording with its captioned video and up to 3 clips, no mark and no sign-up; then the free 7-day Captions + Clips trial, no card (your email only); then Captions + Clips, $29 a month, gives unlimited clips and captioned videos, or Captions, $9.98 a month or $49 once, gives unlimited captioned videos. Captions paid once is not a subscription.
Privacy: the video never leaves your browser. The sound goes only to our speech service, on a phone unless you switch it off, on a computer only if you choose it. Because the video never reaches us, we cannot keep it or train AI on it, and we train no AI on anything you make.
Files: MP4 or MOV in, MP4 out. Captions on the whole video: up to ten minutes. Clips: up to three hours, on a phone or a computer. SRT, VTT, ASS and text files are free without limit for recordings up to ten minutes; a longer recording, up to three hours, gives its script and those files (the first ten minutes free, all of it as one of your free videos, in the trial or with Captions + Clips or Captions) and its clips (every one found and shown before you sign up; up to 3 from each free video exported free, every one in the trial or with Captions + Clips).
Clips: about one clip for every two minutes of talk, up to 30, ranked, the ones we would post first circled, each showing the words it opens and ends on and the reason it was chosen; each cut to 9:16 with ums and dead air removed, an opening card and captions, framed on whoever is talking. Every clip of the whole recording is found and shown, free and with no sign-up; each free video exports up to 3 of them, and the 7-day trial and Captions + Clips export every clip, unlimited.
Languages and styles: captions in the language spoken, twelve Indian languages heard on your device, translation into 59 languages, 60 caption styles.
After it listens: correct any word (one click changes it everywhere), add or cut words, one to six words at a time, position, size, text and highlight colours, and names this browser remembers, all free and without listening again.
Made by: Jordan Labs Ltd, UK company no. 17358646.
Against other tools: sixteen caption tools compared, from the vendors' own pages.