hwrg/end-to-end-tts-fine-tune

Use FastSpeech2 and HiFi-GAN to easily perform end-to-end Korean speech synthesis.

Hi there 🙃

I work at AITRICS Virtual Human Team as a Speech AI Engineer.
I am now working on the engineering and research of Speech Synthesis. (Emotional, Zero-shot, etc..)

News

• 2023.05 : My ZET-Speech paper is accepted by INTERSPEECH 2023 🎉 [Demo]
• 2022.11 : My Few-shot TTS Project paper is accepted as an excellence award by KSC 2022 🎉

Interests

• Speech Synthesis
• Voice Conversion
• Singing Voice Synthesis
• MLOps (Docker, Kubernetes, etc..)
  

Projects

• Develop my own Speech Synthesis model (Please refer to Few-shot TTS Project repository)
  
• Algorithm Study (DP, Graph, etc..)

Blog

https://hwrg.github.io/

Ask me

[email protected]

end-to-end-tts-fine-tune's Issues

Fastspeech2, 보코더 pre-trained ckpt

안녕하세요. 소중한 자료 공유 감사합니다.
이 프로젝트를 통해 TTS 모델을 공부하고 있는데, 요청드릴 부분이 있어 이렇게 의견 남깁니다.

readme에 남겨져 있는 링크가 사라졌는지 파일을 찾을 수가 없는데,
혹시 fastspeech2, 보코더 ckpt 파일을 받을 수 있을까요?