How do I convert audio to text?

Drop an audio or video file onto the box above, pick the spoken language (or leave it on Auto Detect), and press Convert. The text appears as it is produced, so you can start reading before the file has finished.

Do I need an account?

No. Transcription works with no signup at all. A free account is optional and only exists so your transcripts follow you between devices — anything you made beforehand is attached to your account automatically when you sign in.

Is there a time limit on the audio?

No. There is no 30-minute cap and no truncation. A three-hour recording is transcribed end to end, in full, for free.

What file formats can I upload?

MP3, WAV, M4A, FLAC, AAC, OGG and WebM audio, plus MP4 and MOV video — the audio track is extracted automatically. MKV and AVI are not supported yet.

Which languages are supported?

Around 99, including Hindi, Urdu, Bengali, Tamil, Telugu, Marathi, Arabic, Spanish, French, German, Chinese and Japanese. Non-Latin scripts are written in their own script — Hindi comes back in Devanagari, not romanised.

What happens if I pick the wrong language?

We check before committing. The first minute is transcribed with automatic detection, and if it disagrees with your choice you get a warning and can switch — before the rest of the file is processed. If you still want to change afterwards, you can re-transcribe in another language without uploading the file again.

Can I ask questions about my transcript?

Yes. Every transcript has a free AI chat beside it that answers from the recording — summaries, key points, action items, or anything you ask. It answers only from the transcript and says so when something is not in there.

Can I edit the transcript?

Yes. Switch on Edit and click any line to correct it. Changes save immediately and edited lines are marked, so you can see what a human has reviewed.

What formats can I export?

Plain text, text with timestamps, SRT and VTT subtitles, and Word (DOCX). All of them are free, with no watermark and no signup.

What happens to my audio?

Your browser converts the audio before anything is sent, so the original file never leaves your device. The converted audio is stored only so you can re-transcribe in another language, and it is deleted automatically after 24 hours. The transcript itself stays until you delete it.

Is my audio used to train AI models?

No. Your recordings are transcribed and then deleted. They are never used for training and never shared with third parties for that purpose.

It uses a state-of-the-art AI speech recognition model. Accuracy depends on your recording — clear speech with little background noise does very well, while heavy crosstalk, music or a poor microphone will be harder. Selecting the right language matters more than anything else.

Each transcription costs us a fraction of a cent to run. That is cheap enough to give away the features other services charge $10–20 a month for.

Free Audio to Text Converter

Q: Does it identify different speakers?

Not yet. Speaker labelling is the one feature we have deliberately left out for now — it costs several times more to run than the transcription itself, and we would rather keep everything else genuinely free than charge for the whole tool.

Convert audio to text online in seconds. Upload MP3, WAV, M4A or any video file and get an accurate transcript — no 30-minute cap, no signup, every export format free.

No time limit 99 languages Deleted after 24h

Drop your audio or video file here

or click below to browse — no signup, no time limit

Choose audio file

MP3WAVM4AFLACAACOGGMP4MOV

languages: 99 languages
file length: ∞ file length
export formats: 5 export formats
signup steps: 0 signup steps
until auto-delete: 24h until auto-delete

Everything you need

And all of it free — including what most transcription tools put behind a subscription.

No time limit

Three-hour lectures, full podcast episodes, day-long interviews. Nothing is truncated at 30 minutes.

99 languages, proper scripts

Hindi in Devanagari, Arabic in Arabic script — not romanised. Includes Urdu, Bengali, Tamil, Telugu and Marathi.

Language checked first

We detect the language from the first minute and warn you if it disagrees with your choice — before the rest is processed.

Your file never uploads

Your browser converts the audio locally, so the original recording stays on your device.

Five export formats

Plain text, timestamped text, SRT and VTT subtitles, and Word documents. No watermarks.

Editable, with playback

Click any line to correct it. Play the audio back and the transcript follows along.

How it works

Four steps, no account, nothing to download.

1
Drop your file

Audio or video, any length. Nothing to install.
2
Confirm language

Detected from the first minute, checked with you.
3
Watch it appear

Text streams in part by part as it is produced.
4
Edit and export

Correct anything, download in five formats.

See what you get back

Timestamped, punctuated and split into readable lines — in the language's own script. Switch the tabs to see Hindi, Arabic and Spanish output.

Timestamps on every line, clickable back to the audio
Proper Devanagari, Arabic and CJK — never romanised
Editable in place, then exported however you need

Example transcript for illustration — not a real recording.

team-standup.mp3 — transcript

transcribing…

Supported formats

Audio and video. The audio track is extracted automatically.

MP3 WAV M4A FLAC AAC OGG MP4 MOV

No file size limit — your browser compresses the audio before anything is sent. MKV and AVI are not supported yet.

Perfect for everyone

One tool, whatever you are recording.

Students

Turn recorded lectures into searchable notes you can revise from.

Professionals

Meetings and calls converted into text you can skim in seconds.

Creators

Show notes and SRT subtitles produced from a single upload.

Journalists

Quotable interview text with timestamps back to the exact moment.

Researchers

Field recordings and interviews turned into analysable text.

Accessibility

Captions and transcripts so your content works for everyone.

Why is it free?

A fair question. Our costs per recording are a fraction of a cent, which is cheap enough to give away the features other services charge $10–20 a month for — so we do.

Feature	Typical free plan	AudioToText.dev
Free file length	Capped, often 30 minutes	Unlimited
Transcript export	Paid plan	TXT, SRT, VTT, DOCX — free
AI chat over transcript	Top tier only	Free
Editing	Paid plan	Free
Wrong-language protection	None — you find out after	Checked before processing
Re-run in another language	Upload the file again	One click, no re-upload
Signup to try	Usually required	Never

Speaker labelling is the one thing we do not offer — it costs several times more to run than the transcription itself, and we would rather keep everything else free.

Frequently asked questions

Still stuck? Email hello@audiototext.dev and a human will reply.

Ready to convert your audio?

No account, no time limit, no payment details. Just drop a file in.

Start converting — it's free