Phone call recording to text

Permission comes first

Recording rules differ by country and sometimes by state, and in some places everyone on a recording has to agree to it. Check yours before you record, and upload only recordings you have the right to use. Many recorders play an announcement or a tone for exactly this reason. This site cannot tell you what is legal where you are, so if the call matters for a dispute, a contract or a report, ask someone who can.

What the recording is like

A call carries narrowband audio, and some recorders save it at an even lower quality. Voices sound thin and consonants blur, so names, numbers and similar-sounding words are where a transcript goes wrong first. How the call was captured matters as much as the format: a recorder that taps the call itself gives you both sides; an app that listens on the speakerphone gives you a muffled, one-room version of both; and some apps capture only your own side. Do a ten-second test call before an important one.

Steps

1. Find the recording in the recorder app and share or save it as a file (the usual formats are M4A, MP3, AMR and WAV, and all of them upload as they are).
2. Choose it in the box above and set the language if it is not English.
3. Read the transcript with the audio playing. A free account lets you name the speakers, up to eight.
4. Correct names and numbers, then export text, SRT or JSON.

Two voices on separate channels

A stereo call recording sometimes puts you on the left and the other person on the right. The transcriber mixes both into one track, so overlapping speech stays overlapped, and who is speaking comes from the speaker labels you add, not from the channels.

Sales, support and interview calls

The same file works for any kind of call. If the call is a sales or support call, the sales call guide covers what to pull from the transcript, and the phone call recording guide covers quality and consent in more detail. Without an account one file can be up to 15 minutes and 25 MB. A free account lifts those per-file limits and keeps your transcripts in a library, so for a longer recording create one instead of cutting the file up.