Pei-Yin Yang

ReLMCodec: Designing Predictable Speech Tokens from Pre-Quantization Phoneme Structure

ReLMCodec is a low-bitrate single-codebook speech codec built upon a preserve--control--refine principle that moves the empirical single-stream predictability--reconstruction frontier in the evaluations, with gains that carry over to downstream text-to-speech (TTS) synthesis in both intelligibility and speaker similarity.