---
license: mit
language:
- ja
base_model:
- Akjava/matcha-tts_ja_100speakers_group003f-CL-V2
---
# Matcha-TTS Japanese 25 speaker of 100 slots

- contain onnx
- This model has 100 slots and 25 speaker use them

<div class="audio-container">
  <h4>家具商人のフィシェルは、荷車と仔馬を貸してくれた。</h4>
   <h5>spk10:A lower-pitched female voice with a strong core</h5>
  <audio controls src="https://huggingface.co/Akjava/matcha-tts_ja_100speakers_group005qw-025/resolve/main/examples/qwen025_checkpoint_epoch=5749_ch10_kagu.wav"></audio>
  <h4>これはオンクスで作られた音声です。</h4>
   <h5>spk13:A delicate, fleeting female voice that makes you want to protect her</h5>
  <audio controls src="https://huggingface.co/Akjava/matcha-tts_ja_100speakers_group005qw-025/resolve/main/examples/qwen025_checkpoint_epoch=5749_ch13_onnx.wav"></audio>
  <h4>私の声はどうですか？</h4>
   <h5>spk22:A calm, persuasive female voice</h5>
  <audio controls src="https://huggingface.co/Akjava/matcha-tts_ja_100speakers_group005qw-025/resolve/main/examples/qwen025_checkpoint_epoch=5749_ch22_myvoice.wav"></audio>
</div>

## license
My training data is created by Apache Licensed model output.

https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-Base

Matcha-TTS is MIT

https://github.com/shivammehta25/Matcha-TTS
## Inference
see https://huggingface.co/spaces/Akjava/AIGamingVoice-Japanese

## Training
need checkpoint from there

https://huggingface.co/Akjava/matcha-tts_ja_100speakers_group003f-CL-V2

Matcha-TTS-Japanese - contain extra cleaner and jp inference codes

https://github.com/akjava/Matcha-TTS-Japanese

## FAQ
### Want training English
Use official Matcha-TTS

https://github.com/shivammehta25/Matcha-TTS

## Demo
https://ai-game-bu.itch.io/ai-gaming-voice