TokenChain: A Discrete Speech Chain via Semantic Token Modeling
This is an incremental improvement for speech processing systems, enhancing joint ASR and TTS performance through a token-based interface.
The paper tackles the problem of improving automatic speech recognition (ASR) and text-to-speech (TTS) systems by proposing TokenChain, a fully discrete speech chain that couples semantic-token ASR with a two-stage TTS. The result shows TokenChain surpasses baseline accuracy 2-6 epochs earlier, yields 5-13% lower equal-epoch error on LibriSpeech, and reduces relative ASR WER by 56% and T2S WER by 31% on TED-LIUM.
Machine Speech Chain, simulating the human perception-production loop, proves effective in jointly improving ASR and TTS. We propose TokenChain, a fully discrete speech chain coupling semantic-token ASR with a two-stage TTS: an autoregressive text-to-semantic model co-trained with ASR and a masked-generative semantic-to-acoustic model for synthesis only. End-to-end feedback across the text interface is enabled with straight-through argmax/Gumbel-Softmax and balanced with supervised ASR via dynamic weight averaging. Ablations examine optimal temperature schedules for in- and cross-domain transfer. Evaluation reveals TokenChain surpasses baseline accuracy 2-6 epochs earlier and yields 5-13% lower equal-epoch error with stable T2S on LibriSpeech, and reduces relative ASR WER by 56% and T2S WER by 31% on TED-LIUM with minimal forgetting, showing that chain learning remains effective with token interfaces and models.