Language-Statistical Analysis of Neural Audio Codec Tokens Across Architectures, Corpora, and Noise Conditions
This paper analyzes the token statistics of 13 NACs spanning multi-codebook residual vector quantization, single-codebook VQ, and non-VQ designs, evaluated on three corpora under clean, white-noise, and real-world DEMAND-noise conditions to provide architecture-conditioned conventions for applying language-statistical analysis to NAC tokens.
Joonyong Park, Shinnosuke Takamichi, David M. Chan et al.
· 1 citation