pi-audio-feedback

内容来源：README.md（说明文档） · 原始地址 · 查看安装指南

原始内容

![banner-artwork](https://raw.githubusercontent.com/derekluciani/pi-audio-feedback/main/assets/media/banner.webp)

### ▶️ [Watch the demo!](https://youtu.be/HLmf6yvpm7g)

`pi-audio-feedback` is an extension for the Pi Agent Harness (TUI mode) that provides short audio
cues for common Pi lifecycle events.</br>

This can be useful for those wanting non-visual cues, for example when the agent finishes a task
while running in the background.</br>

The extension supports audio cues for the events:

- `piStart`
- `agentStart`
- `agentSettled`
- `agentAborted`
- `toolError`
- `settingsUI`

## Themes

Choose from 4 different audio themes:

- 🔊 **Core**
- 🔊 **Retro**
- 🔊 **Organic**
- 🔊 **Soft**

All sounds are sourced from the [@web-kit/audio](https://audio.raphaelsalaja.com/) library, created
by [@raphaelsalaja](https://github.com/raphaelsalaja).</br>

**Build time:** Each sound patch is a `.json` recipe that's synthesized using the `web audio api`
and captured as a `.wav` file.

## Install

```sh
pi install npm:pi-audio-feedback
```

## Settings

Run the command in Pi:

```text
/audio:config
```

The user can control the playback of any sound (on/off) and choose from 1 of the 4 available audio
themes.

## Compatibility

- **Pi:** >=0.80.6 and <1.0.0
- **Node.js:** >=20
- **Supported local TUI platforms:**
  - macOS 14+ Apple Silicon 🟢 _tested_
  - Windows 10+ x64, PowerShell 5+ 🟢 _tested_
  - Linux Ubuntu/Debian 22+ x64, PipeWire/PulseAudio ⚫️ _not tested_
  - WSL is explicitly unsupported.
- Audio is intentionally unavailable in non-TUI modes, including RPC, JSON, and print mode. Audio is
  also suppressed in CI and SSH sessions. WSL, remote-host and remote-client playback, and other
  non-interactive environments are unsupported.

## License and notices

Package code, generated audio assets, and documentation are licensed under the MIT License. See
[`LICENSE`](https://github.com/derekluciani/pi-audio-feedback/tree/HEAD/LICENSE), which also contains the required third-party notices.

▶️ Watch the demo!

pi-audio-feedback is an extension for the Pi Agent Harness (TUI mode) that provides short audio cues for common Pi lifecycle events.

This can be useful for those wanting non-visual cues, for example when the agent finishes a task while running in the background.

The extension supports audio cues for the events:

piStart
agentStart
agentSettled
agentAborted
toolError
settingsUI

Themes

Choose from 4 different audio themes:

🔊 Core
🔊 Retro
🔊 Organic
🔊 Soft

All sounds are sourced from the @web-kit/audio library, created by @raphaelsalaja.

Build time: Each sound patch is a .json recipe that's synthesized using the web audio api and captured as a .wav file.

Install

pi install npm:pi-audio-feedback

Settings

Run the command in Pi:

/audio:config

The user can control the playback of any sound (on/off) and choose from 1 of the 4 available audio themes.

Compatibility

Pi: >=0.80.6 and <1.0.0
Node.js: >=20
Supported local TUI platforms:
- macOS 14+ Apple Silicon 🟢 tested
- Windows 10+ x64, PowerShell 5+ 🟢 tested
- Linux Ubuntu/Debian 22+ x64, PipeWire/PulseAudio ⚫️ not tested
- WSL is explicitly unsupported.
Audio is intentionally unavailable in non-TUI modes, including RPC, JSON, and print mode. Audio is also suppressed in CI and SSH sessions. WSL, remote-host and remote-client playback, and other non-interactive environments are unsupported.

License and notices

Package code, generated audio assets, and documentation are licensed under the MIT License. See LICENSE, which also contains the required third-party notices.