Vox4ai tts

Skill vox4ai/vox4ai-agent-skills/skills/vox4ai-tts

Text-to-Speech synthesis using vox4ai CLI. Use when the user wants to convert Japanese text to speech, read text aloud, save speech to audio files, list available TTS engines, test TTS connections, or diagnose TTS setup. Supports Edge TTS (cloud), AivisSpeech (local), and other TTS engines via tts-plugin-bridge.From its SKILL.md

Install

npx -y skills add vox4ai/vox4ai-agent-skills --skill vox4ai-tts

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.5 KB, 594 tokens by cl100k_base, as published. Nobody here has run it

vox4ai TTS Agent Skill

Convert Japanese text to speech using vox4ai CLI and tts-plugin-bridge engines.

Quick Start

# Say text aloud (uses configured default engine/model)
vox4ai say "こんにちは"

# Say with specific engine
vox4ai say "こんにちは" -e edgetts

# Specify voice model
vox4ai say "こんにちは" -e edgetts --model ja-JP-KeitaNeural

# Save to file
vox4ai save "こんにちは" -o output.wav

# Save and play after saving
vox4ai save "こんにちは" -o output.wav --play

# List available engines
vox4ai list

# Test engine connection
vox4ai test -e aivisspeech --server-url http://localhost:10101 --style-id 888753760

# Diagnose environment
vox4ai --doctor

# Show current config
vox4ai config

Configuration

Settings are read from ~/.config/vox4ai/config.yaml:

engine: edgetts
model: ja-JP-KeitaNeural
server_url: http://localhost:10101
speed: 1.0
style_id: 888753760

Priority: CLI arguments > config.yaml > built-in defaults

Available TTS Engines

Engines are auto-discovered via entry_points. Common ones:

Engine	Plugin	Type	Notes
edgetts	tts-plugin-edgetts	Cloud (Microsoft Edge TTS)	No server needed
aivisspeech	tts-plugin-aivisspeech	Local	Needs AivisSpeech Engine running

Requirements

Python 3.10+
uv (package manager)
ffplay (for streaming playback) or paplay/aplay (for file playback)

Installation:

uv tool install vox4ai
# or
uv add vox4ai

Troubleshooting

Engine not found

Run vox4ai list to see registered engines. If empty, no plugins are installed.

Connection refused (aivisspeech)

Ensure AivisSpeech Engine is running. Test with:

vox4ai test -e aivisspeech --server-url http://localhost:10101

Voice model not working

Edge TTS Japanese voices: ja-JP-NanamiNeural (female), ja-JP-KeitaNeural (male).

No audio playback

Ensure ffplay (streaming) or paplay/aplay (file) is installed. vox4ai --doctor checks for these.

Long text

For very long texts, use vox4ai save to a file, then play the file.

What ships with it: 3 files

761 B alongside SKILL.md, 3 of them executable

scripts/

tts_list.pyruns233 B
tts_say.pyruns264 B
tts_test.pyruns264 B