JFK — “We choose to go to the Moon” (1962)
Audio · 0:53 · public domain
Separate the speakers in a recording
Upload an audio or video recording with several people and enable speaker separation. Transcrever creates passages with voice labels so you can listen, identify each participant, replace the labels with the correct names, and download the reviewed text.
no sign-uppublic examplesspeaker separationexports TXT, DOCX, SRT, and VTT
We are confirming which transcription options are available.
Galette v. NJ Transit — Supreme Court oral argument (2026)
Audio · 1:35 · 2 speakers · public domain
Who benefits from separating speakers?
Speaker separation helps anyone who needs to know who said each passage in a recorded conversation. It is useful for interviews, meetings, lectures with questions, podcasts with guests, and other files in which voices alternate.
Use the labels to find a remark, confirm its attribution before quoting someone, and keep participant names consistent in the text you share.
How should you prepare a multi-speaker recording?
Choose the file in which everyone can be heard, select the spoken language, and enable speaker separation before uploading. When you control the recording, keep the microphone near the participants and avoid music or noise competing with their voices.
Before uploading, you can add names, companies, and specialist terms as vocabulary hints. Hints guide recognition but do not guarantee spelling, so check important terms against the recording.
If the conversation contains interruptions or people talking at the same time, note the important passages. Check those sections first during review.
What do you receive?
- Editable timed passages made from the speech in the recording.
- A voice label on each passage for following speaker changes.
- Playback of the original file for checking words and identities.
- Search by text or label after you replace the label with the correct name.
- TXT, DOCX, SRT, or VTT download after review.
The labels are a starting point. You confirm who is speaking from the recording and enter the participants' names in the editor.
How to review and rename speakers
1. Listen to a clear passage from each voice
Open a clear passage for each label and use the recording, the participants' introductions, or the conversation context to work out who is speaking.
2. Replace the label with the correct name
Enter editing mode and rename the label. The new name then appears in every passage that uses the same label.
If only one passage is assigned to the wrong person, change the speaker for that passage. You can also remove the label when its speaker is not identified.
3. Correct and organize passages
Use find and replace to correct every exact occurrence of a name or term. When needed, split a passage, merge adjacent passages from the same speaker, delete a passage, or edit its start and end times.
4. Check important speaker changes
Search for a participant's name, open the matching passages, and listen again to anything you plan to quote, publish, or share.
How can you use the reviewed speaker labels?
Download TXT for plain text or DOCX to continue working in a document. Use SRT or VTT when you need timed subtitles. Reviewed labels appear in the files; a passage with no label is exported with its text only.
Saved corrections remain available after a refresh in the same browser. Cross-device task history is available only on Pro and Studio.
A practical review example
For a two-person interview, first listen to one clear question and answer. Rename the two labels as interviewer and interviewee, search for the main subject, and compare every selected quotation with the recording before copying or downloading the text.
When does the recording need more attention?
Overlapping speech, interruptions, echo, noise, or a participant who is far from the microphone can assign a passage to the wrong label or hide words. Listen to speaker transitions and important remarks before attributing the content to someone.
Questions about speaker separation
Does Transcrever know each person's name?
Not automatically. Speaker separation creates labels for the voices; listen to the recording, identify the participants, and replace each label with the correct name.
Can I rename one speaker throughout the transcript?
Yes. When you rename a label in the editor, the new name is applied to every passage that uses that label.
Can I change the speaker for just one passage?
Yes. You can assign another speaker to one passage without changing the other passages that use the label. You can also remove the speaker from that passage when the voice cannot be identified.
What should I check when two people speak at the same time?
Listen to the complete passage and the sentences immediately before and after it. Correct the words and confirm who spoke before using a quotation or attributing a statement.
Use reviewed speakers in interviews and meetings
For an interview, check questions, answers, and quotations. For a meeting, confirm participants and important remarks before sharing the transcript.
Separate the voices before using the text
Upload the recording, enable speaker separation, and check every name and important passage before downloading.
Plans starting at $11.99/month
Pro includes 500 audio minutes per month.
Buy minutes
200 minutes for $6.00
No subscription required. Buy once and use whenever you need.
One-time payment. Minutes never expire.