Codecs

X-Codec2 setup and usage — single-codebook codec used by Llasa for TTS. Stable version is 1.3.0.

WavTokenizer setup — extreme compression (40–75 tok/sec, single codebook). Best for clean TTS; avoid for noisy/degraded speech.

SNAC setup and encode/decode wrapper — matches DAC quality at significantly lower token rate via multi-scale residual vector quantization.

DAC setup and a wrapper class for encoding/decoding audio as flattened token sequences — time-major and codebook-major layouts.

Side-by-side comparison of DAC, SNAC, WavTokenizer, and X-Codec2 — token rates, codebook counts, and when to use each.