Audio to Text

Turn speech in audio or video into text you can copy, or a Word or SRT file.

Processed in your browser — files never leave your device

Drop a file here

or

Select file

About Audio to Text

Turn a recording, voice note, lecture or video into written text in the language that is spoken — Hindi in Devanagari, Tamil in Tamil script, English and about 100 more. Copy the text, or download it as a text file, a Word document or SRT/VTT subtitles, with optional timestamps. Transcription runs in your browser.

How Audio to Text works

Audio to Text uses Whisper, OpenAI's open-source speech recognition model, running in your browser. The sound of your file is first converted to 16 kHz mono with FFmpeg; Whisper then writes what is said, sentence by sentence, with the time each sentence starts. It recognises about 100 languages and writes each in its own script.

Sentences are joined into paragraphs, with a new paragraph after each longer pause. You can download plain text, a Word document, or SRT and VTT subtitle files timed to the speech. Accurate uses Whisper small (about 250 MB, downloaded once); Fast uses Whisper base (about 77 MB) and suits clear English.

When to use Audio to Text

  • Write up an interview, meeting or class recording.
  • Turn a voice note into a message you can copy and edit.
  • Get the script of your own video for a blog post or description.
  • Make subtitles for YouTube from a talk or tutorial.

How to use Audio to Text

  1. Add a recording or video. Drop an MP3, M4A, WAV, MP4 or other audio or video file into the box above.
  2. Choose language and format. Leave the language on Detect automatically or choose it, then pick Text, Word, SRT or VTT.
  3. Convert and copy. Click Convert to text. Read the result on the page, copy it, or download any of the formats.

Options explained

Accurate or Fast
Accurate works for every language, including Hindi and other Indian languages. Fast downloads less and is quicker, but is reliable mainly for English.
Add timestamps
Writes the time before each sentence, like [1:05], so you can find it in the recording.
Word (.docx)
Opens in Microsoft Word, Google Docs and LibreOffice, with Urdu set right to left.
SRT and VTT
Subtitle files timed to the speech, for YouTube, video players and editors.

Frequently asked questions

Can it transcribe a YouTube video?

Download or record the video as a file first, then upload it here. The tool works on files on your device and does not fetch videos from websites.

How accurate is the transcription?

Clear speech with little background noise is transcribed very well. Strong accents, several people talking at once, music and noise lower the accuracy, so read through the text before you use it.

How long can the recording be?

There is no fixed limit, but transcription runs on your device: an hour of audio takes a long time on a phone. A laptop handles a 10-minute file in a few minutes.

Is my recording uploaded?

No. Speech recognition runs in your browser; only the model is downloaded, once.

Is Audio to Text free?

Yes. Audio to Text on DocEpic is free, with no sign-up, no watermark and no limit on how often you use it. The site is supported by ads.

Are my files safe with Audio to Text?

Yes. Audio to Text runs entirely in your web browser, so your files never leave your device and are never uploaded to us.

Do I need to install any software?

No. Audio to Text works in any modern browser on Windows, macOS, Linux, Android and iPhone.

Is there a file size limit?

Files up to 1024 MB are accepted. Because the work happens on your device, very large files depend on your device's memory.

Related tools