Trained Speech Tokenizer to support multiple Asian languages, achieving SOTA results on viVoice and LibriTTS-R