Tokenchain: A Discrete Speech Chain Via Semantic Token Modeling

Abstract

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.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keywang2025tokenchain

Related papers

• Listening While Speaking And Visualizing: Improving ASR Through Multimodal Chain (2019)4.52
• Machine Speech Chain With One-shot Speaker Adaptation (2018)11.29
• Exploring Machine Speech Chain For Domain Adaptation And Few-shot Speaker Adaptation (2021)0.00
• Single-stage TTS With Masked Audio Token Modeling And Semantic Knowledge Distillation (2024)0.00
• End-to-end Feedback Loss In Speech Chain Framework Via Straight-through Estimator (2018)9.03
• High Fidelity Text-to-speech Via Discrete Tokens Using Token Transducer And Group Masked Language Model (2024)4.52
• Tokenverse: Towards Unifying Speech And NLP Tasks Via Transducer-based ASR (2024)1.40
• Internalizing ASR With Implicit Chain Of Thought For Efficient Speech-to-speech Conversational LLM (2024)0.00