Edit a podcast

What it asks you for

Audiorequired

The raw recording.

Remove filler words

Cut ums and ahs.

Add chapters

Derive chapter markers from the transcript.

What it tells the AI to do

The full instructions, exactly as they run.

Edit {{ audio }}.

- Remove filler: {{ removeFiller | default(true) }}
- Chapters: {{ chapters | default(true) }}

Order matters — level LAST, after the cuts, or the loudness target is computed
over audio you then removed.

1. Transcribe for word-level timings.
2. Cut silences over ~0.8s and, if asked, filler words. Leave a short breath;
   cutting to zero makes speech sound clipped.
3. Normalize to −16 LUFS mono / −14 LUFS stereo (the podcast norm), true peak
   under −1 dBTP:

```bash
ffmpeg -i /tmp/input/in.wav -af "loudnorm=I=-16:TP=-1.5:LRA=11" /tmp/output/out.mp3
```

4. Derive chapters from topic shifts in the transcript, not fixed intervals.

Report the before/after duration.
## Delivery

Write the finished artifact into Drive. Never upload a user's content to a
third-party file host (`0x0.st`, `catbox.moe`, `transfer.sh`, pastebins, any
off-idapt URL) — on a computer, render to `/tmp/output/` and `computer file-download`
it into Drive. If you catch yourself reaching for `curl -F`, stop: you already
hold the artifact.

• audio
• podcast
• edit

More audio skills

• Read this aloudTurn text into natural speech with a chosen voice.
• Transcribe and summarizeTurn audio or video into a transcript plus a summary with action items.
• Clean up audio