Use caseUpdated November 4, 20261 min read
Speaker Labels in Transcripts: Who Said What, Without Manual Work
Best for
- Interviews with a host and a guest
- Meetings with several people
- Podcasts with regular co-hosts
- Researchers who quote participants
Not for
- Certified transcripts
- Live calls as they happen
- Video links such as YouTube
How speaker labels work
| Step | What you do | What you get |
|---|---|---|
| 1. Upload | Upload the file, paste a podcast RSS feed, or pass a direct media URL | The audio, ready to transcribe |
| 2. Transcript | Wait a few minutes | Text split by speaker |
| 3. Rename | Replace each generic label with a name | A readable conversation |
| 4. Export | Pick a format | TXT, SRT, VTT, JSON, Word, or PDF |
Speaker labels are automatic, so check them where people talk over each other.
Clear audio helps. Give each person a microphone when you can.
What it costs
Basic is 120 minutes a month free. Pro is $9.99 a month for 1,400 minutes. Packages start at 200 minutes for $2.79 and do not expire. Unused plan minutes do not roll over. English accuracy on the current model is 94.1%. We do not train on your files.
Related
Customer interview transcripts for research. Interview transcript to blog post. Try a file.