Skip to main content

Audio and video to text, in about a minute

Transcribe Audio to Text

Upload an audio or video file and get an editable transcript with speaker labels and timestamps. Rename a speaker once and every line changes. Export Word, PDF, SRT, VTT or an NVivo-ready file. Most files are ready in about a minute. Free for 3 files a day; paid hours never expire.

Drop audio or video here

Interviews, meetings, lectures, podcasts, voice memos

MP3, MP4, M4A, WAV, MOV, WEBM, OGG · up to 5 hours

We never use your files to train models. Files delete after 30 days, or right away when you choose.

Speakers you can rename

Speaker 1 becomes Interviewer everywhere in one edit. Move a single line to another speaker or merge two lines when people talk over each other.

Works on rough audio

Phone calls, noisy rooms and strong accents. Add names and terms before you upload so they come out spelled right.

Pay once, keep the hours

10 hours for $9.99 or 30 hours for $19.99, no subscription and no expiry. Files delete after 30 days, or right away.

How to transcribe audio to text

  1. 1

    Upload or record

    Drop an MP3, MP4, M4A, WAV, MOV or WEBM file, or record on your phone. Video is shrunk to audio in your browser first, so the upload is small.

  2. 2

    Pick the language (optional)

    Language is detected automatically. Choose it yourself for short clips or mixed-language audio, and add names and terms.

  3. 3

    Read the first part while the rest finishes

    The first part shows up in seconds, then the full transcript with speakers. You can close the page; we email you when long files are done.

  4. 4

    Edit, rename speakers, export

    Fix words in place, rename speakers, then export TXT, DOCX, PDF, SRT, VTT or NVivo / MAXQDA format.

What makes a transcript usable

Speaker labels you can fix

No tool gets every speaker change right when people interrupt each other. What matters is how fast you can fix it: rename once, move a line, merge a split sentence.

Timestamps on every paragraph

Researchers and journalists go back to the audio to check a quote. Click any line to play it from that point.

Clean or verbatim

Clean drops um, uh and false starts for reading. Verbatim keeps every word for conversation analysis or legal review. You can switch at any time.

Names and jargon

Speech models misspell rare names. Type them in the names and terms box first; we use them while transcribing and correct close misspellings afterwards.

Price per hour of audio, compared

ServicePlanPricePer hour if you transcribe 20 h/month
Transcribe Audio to Text30 hours, one-time$19.99, never expires$0.67
Transcribe Audio to TextUnlimited$19.99/month or $119.99/year$1.00 (monthly)
TurboScribeUnlimited$20/month, about $10/month yearly$1.00 (monthly)
OtterPro$16.99/month, about $8.33/month yearly$0.85 (monthly)
NottaPro$13.49/month, about $8.17/month yearly$0.67 (monthly)
SonixPay as you go$10 per hour$10.00

Public prices from each pricing page, checked 2026-11-22. Monthly plans shown at their monthly price.

Questions

Is it free to transcribe audio to text here?

Yes, with limits: 3 files a day, up to 30 minutes each, 120 minutes a month, and TXT export. Longer files show the first 30 minutes free and you can unlock the rest. Paid plans add Word, PDF, subtitle and NVivo export.

How accurate is it?

On clear audio most words come out right, including names and numbers. Phone calls, crosstalk and heavy background noise cause more errors, mostly at speaker changes. The editor lets you fix those in place, and adding names to the terms box helps.

How long does it take?

Most files are ready in about a minute; in our tests a 20-minute interview took about 25 seconds and a 55-minute meeting about 70 seconds after upload. The first part appears sooner. Very long files take a few minutes.

Do you use my recordings to train AI?

No. Your files are only used to make your transcript. Audio and text delete automatically after 30 days, after 24 hours if you choose, or right away when you press Delete.

Which languages can I transcribe?

English and about 40 other languages, including Spanish, Portuguese, French, German, Chinese, Japanese, Korean, Hindi and Arabic. The language is detected automatically or you can choose it.

Related

Transcribe your first file now

Free: 3 files a day, first 30 minutes of each. Sign in with Google to start.

Choose a file