Unofficial PyTorch implementation of Higgs Audio V2 Tokenizer with HuBERT semantic features. Complete training pipeline for semantic-acoustic audio tokenization with 960x downsampling and 8-layer RVQ.

pytorch audio-synthesis speech-processing audio-processing vector-quantization dac semantic-features hubert audio-generation neural-audio-codec rvq audio-tokenizer neural-codec higgs-audio speech-tokenization

• Updated Nov 12, 2026
• Python

Toolkit for discrete speech tokenization: SSL feature extraction, k-means units, dedup and unit resynthesis.

pytorch speech-processing self-supervised-learning discrete-units speech-tokenization spoken-language-models

• Updated Jun 28, 2026
• Python