# voicebox-voice-synthesis

Tells your agent how to run Voicebox and talk to its REST API on localhost:17493: speech generation, voice profiles, effects, queue.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 1400 installs · ★ 88
- Author: reason-machines
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/voicebox-voice-synthesis/SKILL.md
- Updated: 2026-10-09
- HTML: https://affpapa.org/en/nevibecode/skills/voicebox-voice-synthesis

## What it does

Voicebox is a local-first TTS studio with voice cloning that runs on your own machine: MLX/Metal on macOS, CUDA on Windows and Linux, CPU as a fallback. The skill hands the agent working recipes for that stack — how to build and start the studio, how to hit `/generate`, `/profiles`, `/generations` and `/models`, how to wire up a TypeScript client, and how to wait on results over SSE.

Inside you get the from-source install steps (`just setup`, `just dev`), a REST reference with `curl` examples, TypeScript wrappers for generation and status polling, and an architecture table covering Tauri, React, FastAPI and SQLite. Five synthesis engines are covered, along with the paralinguistic tags that only `chatterbox-turbo` understands.

Building from source needs Bun, Rust, Python 3.11+ and the Tauri prerequisites, and Linux has no pre-built binary — you compile it yourself. If all you want is speech output, grab a pre-built binary or Docker instead; the skill earns its place when you are digging into the code or the API.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill voicebox-voice-synthesis -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/voicebox-voice-synthesis/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/voicebox-voice-synthesis/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill voicebox-voice-synthesis -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/voicebox-voice-synthesis/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill voicebox-voice-synthesis -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/voicebox-voice-synthesis/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install reason-machines/trending-skills voicebox-voice-synthesis
```

2. **Manually: put the files in place**
- File: `.github/skills/voicebox-voice-synthesis/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill voicebox-voice-synthesis -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/reason-machines/trending-skills --path skills/voicebox-voice-synthesis
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/voicebox-voice-synthesis/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill voicebox-voice-synthesis -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/voicebox-voice-synthesis/SKILL.md

## When to use it
- build Voicebox from source and start the backend with the desktop app
- generate speech from text and a voice profile over REST
- create a profile, upload a sample and export it as an archive
- wait for a generation to finish over SSE instead of polling by hand
- pick a synthesis engine for the job and free GPU memory

## Example prompts
- Build Voicebox from source and run it in dev mode
- POST to /generate with the text and profile_id, then fetch the audio
- Write a TypeScript client that waits for the generation over SSE

## Similar
- [chat-with-anyone](https://affpapa.org/en/nevibecode/skills/chat-with-anyone.md) — skill · 2100 installs
- [elevenlabs](https://affpapa.org/en/nevibecode/skills/elevenlabs.md) — skill · 1200 installs
- [speak-tts](https://affpapa.org/en/nevibecode/skills/speak-tts.md) — skill · 735 installs
- [flyworks-avatar-video](https://affpapa.org/en/nevibecode/skills/flyworks-avatar-video.md) — skill · 374 installs
- [characteristic-voice](https://affpapa.org/en/nevibecode/skills/characteristic-voice.md) — skill · 2800 installs
- [voice-agents](https://affpapa.org/en/nevibecode/skills/sickn33-voice-agents.md) — skill · 890 installs

From the same set: [reason-machines/trending-skills](https://affpapa.org/en/nevibecode/by/reason-machines/trending-skills.md) — 49 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md