SoulX-Duplug: Plug-and-Play Streaming State Prediction Module for Realtime Full-Duplex Speech Conversation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918390119006208 |
|---|---|
| author | Yan, Ruiqi Chen, Wenxi Liu, Zhanxun Ma, Ziyang Lin, Haopeng Wen, Hanlin Xie, Hanke Wu, Jun Liang, Yuzhe Zhao, Yuxiang Feng, Pengchao Qian, Jiale Meng, Hao Dai, Yuhang Yin, Shunshun Tao, Ming Xie, Lei Yu, Kai Wang, Xinsheng Chen, Xie |
| author_facet | Yan, Ruiqi Chen, Wenxi Liu, Zhanxun Ma, Ziyang Lin, Haopeng Wen, Hanlin Xie, Hanke Wu, Jun Liang, Yuzhe Zhao, Yuxiang Feng, Pengchao Qian, Jiale Meng, Hao Dai, Yuhang Yin, Shunshun Tao, Ming Xie, Lei Yu, Kai Wang, Xinsheng Chen, Xie |
| contents | Recent advances in spoken dialogue systems have brought increased attention to human-like full-duplex voice interactions. However, our comprehensive review of this field reveals several challenges, including the difficulty in obtaining training data, catastrophic forgetting, and limited scalability. In this work, we propose SoulX-Duplug, a plug-and-play streaming state prediction module for full-duplex spoken dialogue systems. By jointly performing streaming ASR, SoulX-Duplug explicitly leverages textual information to identify user intent, effectively serving as a semantic VAD. To promote fair evaluation, we introduce SoulX-Duplug-Eval, extending widely used benchmarks with improved bilingual coverage. Experimental results show that SoulX-Duplug enables low-latency streaming dialogue state control, and the system built upon it outperforms existing full-duplex models in overall turn management and latency performance. We have open-sourced SoulX-Duplug and SoulX-Duplug-Eval. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_14877 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SoulX-Duplug: Plug-and-Play Streaming State Prediction Module for Realtime Full-Duplex Speech Conversation Yan, Ruiqi Chen, Wenxi Liu, Zhanxun Ma, Ziyang Lin, Haopeng Wen, Hanlin Xie, Hanke Wu, Jun Liang, Yuzhe Zhao, Yuxiang Feng, Pengchao Qian, Jiale Meng, Hao Dai, Yuhang Yin, Shunshun Tao, Ming Xie, Lei Yu, Kai Wang, Xinsheng Chen, Xie Audio and Speech Processing Recent advances in spoken dialogue systems have brought increased attention to human-like full-duplex voice interactions. However, our comprehensive review of this field reveals several challenges, including the difficulty in obtaining training data, catastrophic forgetting, and limited scalability. In this work, we propose SoulX-Duplug, a plug-and-play streaming state prediction module for full-duplex spoken dialogue systems. By jointly performing streaming ASR, SoulX-Duplug explicitly leverages textual information to identify user intent, effectively serving as a semantic VAD. To promote fair evaluation, we introduce SoulX-Duplug-Eval, extending widely used benchmarks with improved bilingual coverage. Experimental results show that SoulX-Duplug enables low-latency streaming dialogue state control, and the system built upon it outperforms existing full-duplex models in overall turn management and latency performance. We have open-sourced SoulX-Duplug and SoulX-Duplug-Eval. |
| title | SoulX-Duplug: Plug-and-Play Streaming State Prediction Module for Realtime Full-Duplex Speech Conversation |
| topic | Audio and Speech Processing |
| url | https://arxiv.org/abs/2603.14877 |