Back to Models
Kokoro 82M
Text to SpeechKokoro 82M - High-quality TTS with 24 voices, Misaki dictionary
Integration
main.rs
use xybrid_sdk::{Xybrid, Envelope};
// Load the TTS model
let model = Xybrid::model("kokoro-82m").load()?;
// Generate speech from text
let result = model.run(&Envelope::text("Hello from Xybrid!"))?;
// Save the audio output
std::fs::write("output.wav", result.audio_bytes.unwrap())?;Details
- Task
- Text to Speech
- Family
- Hexgrad
- Parameters
- 82M
- Format
- onnx
- Quantization
- fp16
- Size
- 174.9 MB
- Model ID
- kokoro-82m