---
license: mit
language:
- en
tags:
- audio
- text-to-speech
- matcha-tts
---
# Matcha-TTS CommonVoice EN001 

Audio
https://commonvoice.mozilla.org/en/datasets
I called Corps 1 of audios 42da7f26(head-audio-id)_290(files) EN001

Train with IPA(this folk)
https://github.com/akjava/Matcha-TTS-Japanese
## Files
### checkpoints
Matcha-TTS checkpoint - epoch seems big but train with only 290 audios
### ONNX

onnx simplified 
```
from onnxsim import simplify
import onnx

model = onnx.load("en001_6399_T2.onnx")
model_simp, check = simplify(model)

onnx.save(model_simp, "en001_6399_T2_simplify.onnx")
```
- T2 means Vocoder is hifigan_T2_v1
- Unif means Voder is hifigan_univ_v1

To use onnx need something,I'll add sample code later
### Audio
I cut with VAD tools and denoise with resemble-enhance