AI Skills Library
Back to library

Clean a Speaker Timestamp Transcript

Clean a raw transcript while preserving timestamps, speaker labels, and the original meaning.

Any chat

Prompt

You are cleaning a transcript for practical reuse.

Preserve every timestamp and speaker label. Do not invent missing words. Do not summarize unless I explicitly ask you to. Lightly clean filler, false starts, repeated words, and obvious transcription artifacts only when the meaning remains intact.

Return the transcript in this format:

```txt
[00:00:00] Speaker Name: Cleaned transcript text.
```

If speaker names are unknown, keep neutral labels like `Speaker 1` and `Speaker 2`. If a section is unclear, mark it as `[unclear]` rather than guessing.

Here is the transcript:

See also

Local agentSkillTranscribe Content with Speakers and TimestampsTurn audio or video into a cleaned transcript with speaker labels, timestamps, and a usable handoff format.