---
name: audio-trim
description: Trims leading and trailing silence from audio files using FFmpeg. Use when the user wants audio trim, trim silence at start/end, remove leading/trailing silence, batch SFX cleanup, or voice dialogue preprocessing.
---

# Audio Trim

Remove leading/trailing silence via FFmpeg `silenceremove`.

## Rules

When this skill applies, read and follow [skill-dependency-manager](../skill-dependency-manager.md) — run scripts as documented, install missing tools into `.dependency/`.

## Quick Start

Default: threshold **-50 dB**, both sides, output to `trimmed/`:

```bash
.dependency/python/python .cursor/skills/audio-trim/scripts/trim.py path/to/audio_or_folder
```

## Thresholds

| Asset | Threshold |
|-------|-----------|
| UI / SFX / voice | -50 dB (default) |
| Clipped attack / noisy room | -40 to -45 dB |
| Ambience / loops | -60 dB or skip |

## Common Flags

`-t` (threshold dB) · `--no-start` · `--no-end` · `-r` · `-o` / `--output-dir` · `--dry-run` · `--overwrite`

```bash
.dependency/python/python .cursor/skills/audio-trim/scripts/trim.py Audio/SFX -t -50 --no-end -r --dry-run
```

Originals are never modified. Supported: `.wav`, `.mp3`, `.ogg`, `.flac`, `.aac`, `.m4a`, `.wma`.

## Agent Notes

1. Use the bundled script, not ad-hoc `-ss`/`-to`.
2. **Looping BGM** — avoid batch end-trim; can break loop seams.
3. Attack or reverb cut off → raise `-t` or use `--no-end`.
4. FFmpeg filter details: [reference.md](reference.md)