Skip to content
Sondeas
Start 7 days free

Analytics only if you say yes

Essential cookies run the site. With your yes we also measure visits and ad results. Nothing optional loads until you choose.

No recording or note content is ever sent to analytics.

Details in our privacy policy. Change your mind any time from Privacy choices in the footer.

Audio transcription

Audio to text converter

Upload MP3, M4A, WAV or MP4, even recordings hours long. Get a timestamped transcript with speakers, then edit, translate or export it.

  • 7 days free on Pro and Clinician
  • Cancel any time
  • Speakers and timestamps
  • 100 languages

Pay as you go: $3 an hour of audio. The summary, translation and exports are included.

Transcribe a file

7 days free

Next you create an account and choose a plan or pack to get your transcript. Your file waits in this browser; nothing is uploaded before that.

No file at hand?Start 7 days free and record in the browser

How it works

How to convert audio to text

Three steps. You choose the file; Sondeas does the typing.

  1. Step 1

    Drop your file

    Choose an MP3, M4A, WAV, MP4, WebM or OGG file, or drag it onto the box. Pick the spoken language or let Sondeas detect it.

  2. Step 2

    Create your account

    Sign up with Google or email. Your file waits in this browser until you do, then uploads to your private workspace.

  3. Step 3

    Read, fix and export

    Get a timestamped transcript with speakers. Fix a word in place, then summarise, translate or export it.

What you get

A transcript you can work with, not a wall of text

  • Who said what: every line labelled with its speaker, and speakers renamed in one place.
  • A timestamp on every line. Click one to hear that moment.
  • Fix a word in place and search the whole transcript.
  • Turn it into a summary, action items, an email draft or a translation, included in the price.
  • Subtitles for video: export SRT or VTT.
Export asMDTXTDOCXPDFSRTVTTJSON

Formats and limits

Supported formats, limits and tips

Formats and limits of the audio to text converter
File typesMP3, M4A, WAV, MP4, WebM and OGG, plus AAC and FLAC. From a video, the audio track is transcribed.
File sizeUp to 2 GB per file
LengthMulti-hour recordings are fine
Languages100 languages, detected automatically or chosen by you
SpeakersLabelled automatically; rename them in the note
ExportsMD, TXT, DOCX, PDF, SRT, VTT, JSON
Cost$3 an hour of audio. 7 days free on Pro and Clinician. Failed jobs use nothing.
PrivacyEncrypted. Optional passphrase lock. Delete the audio or the note any time.

Tips for the best transcript

  • Record in a quiet room, close to the microphone.
  • With several speakers, one voice at a time gives the cleanest speaker labels.
  • Upload the original file rather than a screen recording of it.

Use cases

What people convert to text

The same account does all of it. Pick the closest match to see what comes after the transcript.

Pricing

$3 an hour of audio · 7 days free on Pro and Clinician

A one-hour interview costs $3. The summary, translation and every export are included. Or try Pro or Clinician free for 7 days.

Start 7 days freeEstimate my costSee all plans

Questions

Audio to text: questions

Anything else? The help centre covers uploads, languages and exports.

Is there a free trial?
For 7 days on Pro or Clinician, with 5 fast hours included. A card is required and you can cancel any time. Without a plan it is $3 an hour of audio, and the summaries, translations and exports you make from it are included.
Why do I need an account?
So your transcript is saved privately to you and you can come back to it. Your file waits in this browser while you sign up; nothing is uploaded before that.
Which file formats can I upload?
MP3, M4A, WAV, MP4, WebM and OGG, plus AAC and FLAC, up to 2 GB per file. From a video, the audio track is transcribed.
How long can the recording be?
Multi-hour recordings are fine: a full meeting, lecture or interview in one file. You only pay for the minutes of audio.
Can it tell speakers apart?
Yes. Each line is labelled with its speaker and a timestamp, and you can rename the speakers in the note.
How accurate is it?
It depends on the recording: clear audio with one voice at a time works best. Every line is linked to its moment in the audio, so you can listen and fix a word in place.
Can I translate the transcript or make subtitles?
Yes. Translate into any of 100 languages with the original kept alongside, and export the transcript as SRT or VTT subtitles. Both are included in the price of the audio.
Who can see my audio and transcript?
Your recordings and notes are private to your account and encrypted. Add a passphrase and only you hold the key. Delete the audio or the note any time.

Keep exploring

Related tools and guides

Your recording, as text you can use.

Drop a file, create your account, read the transcript. 7 days free on Pro and Clinician. Card required, cancel any time.

Start 7 days freeOr go back up and drop your file