---
base_model:
- hexgrad/Kokoro-82M
pipeline_tag: text-to-speech
tags:
- onnx
---
# Kokoro 1.0 [ONNX]

ONNX files exported from [Kokoro 1.0](https://huggingface.co/hexgrad/Kokoro-82M). Voices have been converted to HDF5 and the `speed` input now uses type `float32` (originally `int64`).