vllm_omni.model_executor.models.audio8_tts ¶

Audio8 TTS Preview (DualAR, 44.1 kHz codec) model implementation.

Modules:

Name	Description
`audio8_tts_codec_decoder`	Audio8 TTS Preview -- codec decoder (Stage 1).
`audio8_tts_fast_ar`	Audio8 TTS Fast AR: the residual-codebook predictor.
`audio8_tts_slow_ar`	Audio8 TTS Preview -- Slow AR model (Stage 0).
`codec`	Audio8 TTS neural audio codec (44.1 kHz, 10 codebooks).
`codec_utils`	Loading / caching helpers for the Audio8 TTS `codec.pth` checkpoint.
`configuration_audio8_tts`	Configuration classes for Audio8 TTS Preview (`model_type = "arktts"`).
`pipeline`	Audio8 TTS Preview pipeline topology.
`prompt_utils`	Audio8 TTS prompt construction.
`sampling`	Audio8 TTS sampling primitives, ported 1:1 from the reference checkpoint.