Seed-ASR: Understanding Diverse Speech and Contexts with LLM-based Speech Recognition
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929415575830528 |
|---|---|
| author | Bai, Ye Chen, Jingping Chen, Jitong Chen, Wei Chen, Zhuo Ding, Chuang Dong, Linhao Dong, Qianqian Du, Yujiao Gao, Kepan Gao, Lu Guo, Yi Han, Minglun Han, Ting Hu, Wenchao Hu, Xinying Hu, Yuxiang Hua, Deyu Huang, Lu Huang, Mingkun Huang, Youjia Jin, Jishuo Kong, Fanliu Lan, Zongwei Li, Tianyu Li, Xiaoyang Li, Zeyang Lin, Zehua Liu, Rui Liu, Shouda Lu, Lu Lu, Yizhou Ma, Jingting Ma, Shengtao Pei, Yulin Shen, Chen Tan, Tian Tian, Xiaogang Tu, Ming Wang, Bo Wang, Hao Wang, Yuping Wang, Yuxuan Xia, Hanzhang Xia, Rui Xie, Shuangyi Xu, Hongmin Yang, Meng Zhang, Bihong Zhang, Jun Zhang, Wanyi Zhang, Yang Zhang, Yawei Zheng, Yijie Zou, Ming |
| author_facet | Bai, Ye Chen, Jingping Chen, Jitong Chen, Wei Chen, Zhuo Ding, Chuang Dong, Linhao Dong, Qianqian Du, Yujiao Gao, Kepan Gao, Lu Guo, Yi Han, Minglun Han, Ting Hu, Wenchao Hu, Xinying Hu, Yuxiang Hua, Deyu Huang, Lu Huang, Mingkun Huang, Youjia Jin, Jishuo Kong, Fanliu Lan, Zongwei Li, Tianyu Li, Xiaoyang Li, Zeyang Lin, Zehua Liu, Rui Liu, Shouda Lu, Lu Lu, Yizhou Ma, Jingting Ma, Shengtao Pei, Yulin Shen, Chen Tan, Tian Tian, Xiaogang Tu, Ming Wang, Bo Wang, Hao Wang, Yuping Wang, Yuxuan Xia, Hanzhang Xia, Rui Xie, Shuangyi Xu, Hongmin Yang, Meng Zhang, Bihong Zhang, Jun Zhang, Wanyi Zhang, Yang Zhang, Yawei Zheng, Yijie Zou, Ming |
| contents | Modern automatic speech recognition (ASR) model is required to accurately transcribe diverse speech signals (from different domains, languages, accents, etc) given the specific contextual information in various application scenarios. Classic end-to-end models fused with extra language models perform well, but mainly in data matching scenarios and are gradually approaching a bottleneck. In this work, we introduce Seed-ASR, a large language model (LLM) based speech recognition model. Seed-ASR is developed based on the framework of audio conditioned LLM (AcLLM), leveraging the capabilities of LLMs by inputting continuous speech representations together with contextual information into the LLM. Through stage-wise large-scale training and the elicitation of context-aware capabilities in LLM, Seed-ASR demonstrates significant improvement over end-to-end models on comprehensive evaluation sets, including multiple domains, accents/dialects and languages. Additionally, Seed-ASR can be further deployed to support specific needs in various scenarios without requiring extra language models. Compared to recently released large ASR models, Seed-ASR achieves 10%-40% reduction in word (or character, for Chinese) error rates on Chinese and English public test sets, further demonstrating its powerful performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_04675 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Seed-ASR: Understanding Diverse Speech and Contexts with LLM-based Speech Recognition Bai, Ye Chen, Jingping Chen, Jitong Chen, Wei Chen, Zhuo Ding, Chuang Dong, Linhao Dong, Qianqian Du, Yujiao Gao, Kepan Gao, Lu Guo, Yi Han, Minglun Han, Ting Hu, Wenchao Hu, Xinying Hu, Yuxiang Hua, Deyu Huang, Lu Huang, Mingkun Huang, Youjia Jin, Jishuo Kong, Fanliu Lan, Zongwei Li, Tianyu Li, Xiaoyang Li, Zeyang Lin, Zehua Liu, Rui Liu, Shouda Lu, Lu Lu, Yizhou Ma, Jingting Ma, Shengtao Pei, Yulin Shen, Chen Tan, Tian Tian, Xiaogang Tu, Ming Wang, Bo Wang, Hao Wang, Yuping Wang, Yuxuan Xia, Hanzhang Xia, Rui Xie, Shuangyi Xu, Hongmin Yang, Meng Zhang, Bihong Zhang, Jun Zhang, Wanyi Zhang, Yang Zhang, Yawei Zheng, Yijie Zou, Ming Audio and Speech Processing Sound Modern automatic speech recognition (ASR) model is required to accurately transcribe diverse speech signals (from different domains, languages, accents, etc) given the specific contextual information in various application scenarios. Classic end-to-end models fused with extra language models perform well, but mainly in data matching scenarios and are gradually approaching a bottleneck. In this work, we introduce Seed-ASR, a large language model (LLM) based speech recognition model. Seed-ASR is developed based on the framework of audio conditioned LLM (AcLLM), leveraging the capabilities of LLMs by inputting continuous speech representations together with contextual information into the LLM. Through stage-wise large-scale training and the elicitation of context-aware capabilities in LLM, Seed-ASR demonstrates significant improvement over end-to-end models on comprehensive evaluation sets, including multiple domains, accents/dialects and languages. Additionally, Seed-ASR can be further deployed to support specific needs in various scenarios without requiring extra language models. Compared to recently released large ASR models, Seed-ASR achieves 10%-40% reduction in word (or character, for Chinese) error rates on Chinese and English public test sets, further demonstrating its powerful performance. |
| title | Seed-ASR: Understanding Diverse Speech and Contexts with LLM-based Speech Recognition |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2407.04675 |