Skip to content
Original Voice Soundboard

Getting Started

Get from zero to generated speech in about 60 seconds.

Requirements

Section titled “Requirements”
  • Python 3.10 or newer
  • Approximately 350 MB of disk space for Kokoro models
  • Optional: CUDA-compatible GPU for faster generation

Install

Section titled “Install”
Terminal window
pip install voice-soundboard

This installs the core library with the Kokoro TTS engine. For optional engines and extras, see the Engines page.

Download models

Section titled “Download models”

Kokoro needs two model files. Create a models/ directory and download them:

Terminal window
mkdir models && cd models
curl -LO https://github.com/thewh1teagle/kokoro-onnx/releases/download/model-files-v1.0/kokoro-v1.0.onnx
curl -LO https://github.com/thewh1teagle/kokoro-onnx/releases/download/model-files-v1.0/voices-v1.0.bin

Generate your first speech

Section titled “Generate your first speech”
from voice_soundboard import VoiceEngine


engine = VoiceEngine()
result = engine.speak("Hello world!")
print(result.audio_path)  # output/af_bella_<hash>.wav

That produces a .wav file you can play with any audio player.

CLI usage

Section titled “CLI usage”

If you prefer the command line:

Terminal window
voice-soundboard speak "Hello world!"
# -> output/hello_world.wav

More CLI examples

Section titled “More CLI examples”
Terminal window
# Pick a specific voice
voice-soundboard speak "Cheerio!" --voice bm_george


# Use a preset
voice-soundboard speak "Breaking news!" --preset announcer


# Set an emotion
voice-soundboard speak "I'm so happy!" --emotion excited

What happens under the hood

Section titled “What happens under the hood”
  1. Text goes into the Kokoro ONNX model running locally on your machine
  2. The model generates raw audio samples
  3. Voice Soundboard writes a .wav file to the output/ directory
  4. You get back a result object with the file path and metadata

No network calls. No API keys. No cloud. Everything runs on your hardware.

Next steps

Section titled “Next steps”