Free tool · No sign-up · No upload

Free audio to text, without uploading your file.

Choose a recording, an interview, a lecture or a video. It is transcribed right here in your browser, and you get text you can edit plus SRT or VTT subtitles with real times.

Runs on your device

Drop an audio or video file here

MP3, M4A, WAV, OGG, FLAC, MP4, MOV or WebM. Up to 60 minutes and 300 MB.

Choose an audio or video file. It stays on this computer.

English model. Subtitle files use the text as recognized; edits in the box apply to the .txt and copy. The file and the text stay in this tab.

How it works

Three steps, all in this tab.

1
Choose a file.
Drop it on the box or pick it from your computer. Your browser reads its sound track; nothing is sent anywhere.
2
Let it transcribe.
The first time, the browser downloads a Whisper model (about 80 MB, or about 165 MB where your browser uses the graphics chip). Then the file is recognized part by part, and the text grows as it goes.
3
Edit and save.
Fix names or jargon in the text, then copy it or save a .txt file. Save .srt or .vtt subtitles for your video.

Why it is different

Transcription that never sees a server.

Most free transcription websites upload your file and cap the minutes you get. This page runs Whisper, OpenAI’s open speech recognition model, on your own computer, so an interview, a meeting or a voice memo stays yours.

Private by design.

The file and its transcript stay in this tab. There is no account, no upload and no copy kept by us.

Subtitles with real times.

The file is cut at natural pauses into parts of a few seconds, so each subtitle starts and ends where it was said.

English, or almost any language.

English mode is the most accurate for English. Any language mode recognizes 99 languages and detects which one your file is in, with lower accuracy.

Honest about its limits.

A model small enough for a browser makes more mistakes than large ones, especially with names, crosstalk, music and noisy rooms. It does not tell speakers apart.

Questions.

Is this audio to text converter really free?: Yes. There is no account, no trial and no limit on how many files you transcribe. Each file can be up to 60 minutes long and 300 MB. Yap’s paid product is its desktop dictation app; this page is free on its own.
Is my file uploaded?: No. The file is read and transcribed inside this browser tab, on your computer. The audio and the transcript are not sent to a server or saved by us. The site records usage events, such as that a transcription finished, never the file, its name or its words.
Which files work?: Anything your browser can play: MP3, M4A, WAV, OGG, Opus, FLAC and AAC audio, and the sound track of MP4, MOV and WebM videos. Chrome and Edge read the most formats.
How long does it take?: It depends on your computer. Where the browser lets the model use the graphics chip (current Chrome and Edge), an hour of audio usually takes a few minutes. On the processor alone it can take about as long as the recording. Text appears as each part is done, and you can stop at any time.
Can I make subtitles?: Yes. The file is cut into short parts at natural pauses, and each part becomes a subtitle with its real start and end time. Download them as an .srt or .vtt file for YouTube, video editors and players.
Which languages does it understand?: English mode uses Whisper’s English model, the most accurate choice for English. Any language mode uses the multilingual model, which recognizes 99 languages and detects the file’s language from its first speech. It transcribes in the spoken language and never translates.
Why does it download a model first?: Recognizing speech on your device needs a speech model on your device. The first time, your browser downloads a Whisper model (about 80 MB, or about 165 MB where your browser uses the graphics chip) and keeps it cached for next time, also for the free online dictation page.

Want to speak rather than upload?

In the browser: free online dictation, the same private speech recognition for your microphone.
On a Mac: how to use the Mac’s built-in voice to text, and dictation that works offline.
On Windows: how to dictate on Windows 11 and 10.

Three steps, all in this tab.

Choose a file.

Drop it on the box or pick it from your computer. Your browser reads its sound track; nothing is sent anywhere.

Let it transcribe.

The first time, the browser downloads a Whisper model (about 80 MB, or about 165 MB where your browser uses the graphics chip). Then the file is recognized part by part, and the text grows as it goes.

Edit and save.

Fix names or jargon in the text, then copy it or save a .txt file. Save .srt or .vtt subtitles for your video.

Questions.

Is this audio to text converter really free?

Yes. There is no account, no trial and no limit on how many files you transcribe. Each file can be up to 60 minutes long and 300 MB. Yap’s paid product is its desktop dictation app; this page is free on its own.

Is my file uploaded?

No. The file is read and transcribed inside this browser tab, on your computer. The audio and the transcript are not sent to a server or saved by us. The site records usage events, such as that a transcription finished, never the file, its name or its words.

Which files work?

Anything your browser can play: MP3, M4A, WAV, OGG, Opus, FLAC and AAC audio, and the sound track of MP4, MOV and WebM videos. Chrome and Edge read the most formats.

How long does it take?

It depends on your computer. Where the browser lets the model use the graphics chip (current Chrome and Edge), an hour of audio usually takes a few minutes. On the processor alone it can take about as long as the recording. Text appears as each part is done, and you can stop at any time.

Can I make subtitles?

Yes. The file is cut into short parts at natural pauses, and each part becomes a subtitle with its real start and end time. Download them as an .srt or .vtt file for YouTube, video editors and players.

Which languages does it understand?

English mode uses Whisper’s English model, the most accurate choice for English. Any language mode uses the multilingual model, which recognizes 99 languages and detects the file’s language from its first speech. It transcribes in the spoken language and never translates.

Why does it download a model first?

Recognizing speech on your device needs a speech model on your device. The first time, your browser downloads a Whisper model (about 80 MB, or about 165 MB where your browser uses the graphics chip) and keeps it cached for next time, also for the free online dictation page.