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