What are speaker labels? They are names or numbers such as Speaker 1 that mark who says each part of a transcript. A label stays the same for one voice from start to end, so a reader can follow a conversation. The labels are tags on the text and not a checked identity.
Transcribe a recording
Send your voice message, audio or video to @potok_now_bot on Telegram and the text comes back in the chat.
Open @potok_now_bot in Telegram
Files up to 2 GB · export to TXT, MD, DOCX, PDF, SRT and VTT · speaker labels and timestamps · about 30 languages
Key facts
- Software groups the audio by voice and numbers the groups in order of appearance; it does not know who the people are.
- Two similar voices, or one person on a bad line, can be split or merged wrongly, so a label may switch mid-speech.
- Overlapping speech is the weak point: when two people talk at once, the words are often given to one of them.
- Replacing Speaker 1 and Speaker 2 with real names is a manual find-and-replace that takes a minute once you know who spoke first.
- Labels cost nothing to ignore, so a one-person recording such as a lecture or voice memo is cleaner without them.
What it looks like
Speaker 1: Thanks for joining. Shall we start with the budget?
Speaker 2: Yes. The draft is ready, I only need approval.
Speaker 1: Good. Send it today.
// after a manual find-and-replace
Chair: Thanks for joining. Shall we start with the budget?
Finance: Yes. The draft is ready, I only need approval.
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
Do speaker labels show real names?
No. They are numbers such as Speaker 1 and Speaker 2. Add the real names yourself after reading the transcript.
Why does the same person get two labels?
Changes in distance, microphone or room can make one voice sound like two. Merge them by hand with a find-and-replace.
How do I turn labels on in the Potok bot?
Speaker labels are optional and can be switched on before you send the file. They appear as numbers next to each part of the text.
Which recordings suit labels best?
Interviews, meetings and podcasts with a few people who take turns. Close microphones and little overlap give the cleanest result.
Is a label the same as diarisation?
Diarisation is the process that decides who spoke when. The label is the visible result of it, printed in the transcript.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- Speaker diarisation (Wikipedia), Wikipedia · checked
- Speaker recognition (Wikipedia), Wikipedia · checked
Try it on your own recording
Send your voice message, audio or video to @potok_now_bot on Telegram and the text comes back in the chat.
Open @potok_now_bot in Telegram