1 2 3 4 5 6 7 8 9 10 11 12
//! Qwen3-TTS-12Hz model implementation. //! //! Architecture: Talker LM + Code Predictor → Speech Tokenizer Decoder → 24kHz audio //! Discrete multi-codebook language model approach. pub mod code_predictor; pub mod config; pub mod model; pub mod speech_tokenizer; pub mod talker; pub use model::Qwen3TtsModel;