What is Speaker Separation?
Speaker Separation, also called speaker diarization, identifies different voices and labels transcript segments so you can see who said what during multilingual conversations.
도구
AI Speaker Separation for Meeting Transcripts
베타
Automatically identify who said what during multilingual conversations, then rename or reassign speakers for clearer meeting records.
Know who said what in multilingual meetings
Speaker Separation automatically organizes conversations by speaker, so teams can follow discussions more easily during live translation and review cleaner records afterward.
Automatic identification
Automatically identify different speakers
켜기 Diarization (Beta) from the translation panel before the conversation starts. Transync AI then labels different speakers automatically with clear, color-coded names such as Speaker 1, Speaker 2, and Speaker 3.
Speaker labels appear during live translation, remain available in the saved record, and are also shown in the AI-generated meeting summary, making it easier to see who contributed each key point.
Clearer records
Rename speakers for easier review
Replace generic labels with a person’s name or role, such as Jessica, Host, Interviewer, or Customer. A renamed speaker is updated across the meeting record, making long discussions easier to scan and share.
You can also adjust speaker colors to make individual voices easier to distinguish.
Flexible correction
Reassign a transcript segment when needed
If a transcript segment is assigned to the wrong person, open Change Speaker and choose the correct speaker. This gives you a practical way to clean up the record without rewriting the conversation.
Review speaker assignments before sharing records or using them in meeting follow-up workflows.
Designed for conversations with multiple voices
자주 묻는 질문
What is AI Speaker Separation?
AI Speaker Separation, also known as speaker diarization, identifies different voices in a conversation and labels transcript segments by speaker.
How do I enable Speaker Separation?
Open the translation panel before starting a task and turn on Diarization (Beta). The option is available when you use a supported Monsoon model.
Can I rename a speaker?
Yes. Open the speaker menu and rename a generic label such as Speaker 1 to a person’s name or role. The new name is applied across the meeting record.
Can I correct a speaker assignment?
Yes. Select the transcript segment, open Change Speaker, and assign it to the correct speaker.
Where do speaker labels appear?
Speaker labels can appear during live translation and in the saved translation record, so you can follow the discussion in real time and organize it afterward.
Why is Speaker Separation marked as Beta?
The feature is still being improved. Speaker assignments should be reviewed when recordings contain overlapping speech, background noise, or similar voices.
빠른 답변
Quick answers about Speaker Separation
Short answers about speaker labels, diarization, transcript editing, saved records, and Beta accuracy.
가장 좋은
Best for team meetings, interviews, classes, lectures, and customer conversations where several people speak and the record needs clear speaker attribution.
작동 방식
Turn on Diarization (Beta) before starting a supported translation task. Transync AI automatically assigns color-coded labels such as Speaker 1 and Speaker 2.
During live translation
Speaker labels can appear while people are speaking, making it easier to follow multilingual discussions with several voices in real time.
In saved records
The labels remain in the saved translation record so you can review contributions, decisions, questions, and follow-up items after the conversation.
Rename and correct speakers
You can replace generic labels with names or roles, adjust speaker colors, and use Change Speaker to reassign a transcript segment when needed.
Beta accuracy
Speaker Separation is still being improved. Review assignments when conversations include overlapping speech, background noise, or similar-sounding voices.