Transcribe an audio recording
Choose a file from a voice recorder or recording app and AI (Whisper or SenseVoice) transcribes it inside your browser. There is no need to convert the format, and there is no time limit however long the recording is. The audio never reaches a server. No file? You can record from the microphone on this page and transcribe it right away (for up to 3 hours, or less depending on your device's memory and the selected model).
You can normally leave this on Auto. Change it only if the result clearly has the wrong number of speakers
No file? Record with your microphone. Transcription starts after you stop recording (up to about 000 min on this device with the selected model). Your recording is not sent to any server either.
How to transcribe a voice recorder file
- Connect the voice recorder to your computer over USB and copy the recording. For recordings made with a smartphone app, save the file to cloud storage from the share menu. For iPhone Voice Memos, see the steps on the M4A transcription page.
- If two or more people are speaking, turn on speaker diarization. If you know the language, choose it instead of auto-detect.
- Choose the recording. Finalized lines appear on screen as it goes, and when it finishes you can choose TXT, Word, SRT or another format under "Download".
How to record with the microphone and transcribe
- Press “Record with microphone” and allow microphone access when the browser asks.
- When you finish speaking, press “Stop recording”. The recording limit is shown on the screen while recording. On a phone, locking the screen or switching to another app while recording may stop the recording.
- Check the language, model and speaker diarization, then press “Start transcription”. Press “Save as audio file” to save the recording as a file; it is also kept in this device's history together with the transcript.
File formats by recording device
Many voice recorders save MP3 or WAV, and some save WMA. Smartphone recording apps save M4A, MP3, AMR or other formats. All of these load as they are, so there is no need to convert them to MP3 first.
What affects the result most is how the recording was made. Three things matter: the speaker being close to the microphone, little background noise such as air conditioning, and voices not overlapping.
There is no time limit, even for long recordings. Finalized lines appear while processing, so you can check the content before it finishes. If you close the page partway, the text finalized up to that point stays in your history.
If something goes wrong
- If quiet or distant voices are missed in languages other than Japanese, Chinese and Korean, switch to Small or a larger model. For Spanish, German and other European languages on a computer with WebGPU, Parakeet Ultra is more accurate and faster than Small.
- In long recordings with silence, sentences nobody said may appear. Delete those lines on the results screen.
Frequently asked questions
Which recording formats are supported?
MP3, WAV, M4A, AAC, FLAC, OGG, Opus, AIFF, WMA, AMR and WebM are supported. If you pick a video file, only the audio track is extracted and processed.
How many hours of recording can be transcribed?
Audio files have no time or size limit. How long a recording can be processed at once depends on your device's memory and the selected model. When you record with the microphone on this page, you can record for up to 3 hours, or less depending on your device's memory and the selected model; the limit is shown before you start recording. Phones run in a reduced-memory mode to protect the device, so a computer is recommended for long recordings.
Can it separate multiple speakers in a recording?
Yes. With speaker diarization on, lines are split by speaker (up to eight). For meeting recordings, the Meeting transcription page, which has diarization turned on from the start, is the easiest to use.
Is my recording sent anywhere?
No. Transcription runs in your browser on your device, and neither the recording nor the resulting text is passed to an external server. Audio recorded with the microphone is handled the same way. Processing statistics used to fix problems (device specs, error type and similar) are sent to this site's own server, but they contain no audio, transcript text, file names or IP addresses. You can stop this by turning off the switch under “Help us improve the service” at the bottom of the page.
Getting the audio ready, and after transcribing
Help us improve the service
Sending us the circumstances of a run — successful or failed — helps us track down problems (a failure rate needs both). Knowing which devices and settings fail most often lets us revise the recommended defaults and fix the bugs behind them. If you don't mind, leaving this on is a real help.
What is sent
Your device specs (GPU model, memory, core count), browser and OS, the model and language you chose, your speaker diarization setting (on or off, and the number of speakers you set), the page's display language, how far the run got (including when you stopped it or closed the page), the error type and details, the file's format with a rough size and length, and how long it took. A temporary number that links runs in the same tab (it is deleted when you close the tab). On the result screen: how many times you used editing (such as renaming speakers or splitting lines), translation, summaries and subtitle imports, the formats you chose under “Download”, how often you used “Copy text” and “Download all”, and the translation language pair.
What is never sent
Audio, transcripts (including speaker names and translations), file names, and IP addresses. The data goes only to this site's own server, never to a third-party service.