<< All versions
Skill v1.0.0
Trusted Publisher100/100google-gemini/gemini-managed-agents-templates/audio-mixing
──Details
PublishedMay 19, 2026 at 10:30 PM
Content Hashsha256:706bd4d42fd1d508...
Git SHA
──Files
Files (1 file, 1.5 KB)
SKILL.md1.5 KBactive
SKILL.md · 61 lines · 1.5 KB
version: "1.0.0" name: audio-mixing description: Mix speech audio and background music into a polished radio show file.
Audio Mixing
Combine the TTS speech audio and Lyria background music into a single, polished radio show file.
Embedded Script
bash
python3 skills/audio-mixing/scripts/mix_audio.py --workspace ./workspace
Arguments
| Argument | Default | Description | |
|---|---|---|---|
--workspace | workspace | Root workspace directory |
What it does
- Loads speech from
{workspace}/audio/speech/speech.wav. - Adds 3 seconds of silence padding to the end of the speech to prevent it from being faded out.
- Loads background music from
{workspace}/audio/music/background.mp3(if exists). - Loops music to match speech duration, lowers volume to -18dB.
- Overlays speech on music with a 1-second music intro.
- Adds fade-in/fade-out.
- Exports as MP3.
Dependencies
pydubffmpeg(system)
Mixing Guidelines
| Element | Level | Notes | |
|---|---|---|---|
| Speech | 0 dB | Untouched, full volume | |
| Background music | -18 dB | Barely audible — subtle bed under speech |
Transitions
- Music fade-in: 3 seconds
- Music fade-out: 5 seconds
- Overall fade-in: 500ms
- Overall fade-out: 2 seconds
Output
| File | Path | Format | |
|---|---|---|---|
| MP3 (distribution) | {workspace}/audio/final/ai_radio.mp3 | MP3, 192kbps |
Fallback
If no background music exists, produces speech-only output with just fade-in/fade-out applied.