Year: 2025
Type: preprint
Abstract: In recent years, automatic speech recognition (ASR) has witnessed transformative advancements driven by three complementary paradigms: data scaling, model size scaling, and deep integration with large language models (LLMs). However, LLMs are prone to hallucination, which can significantly degrade user experience in real-world ASR applications. In this paper, we present Fun-ASR, a large-scale, LLM-based ASR system that synergistically combines massive data, large model capacity, LLM integration, and reinforcement learning to achieve state-of-the-art performance across diverse and complex speech recognition scenarios. Moreover, Fun-ASR is specifically optimized for practical deployment, with enhancements in streaming capability, noise robustness, code-switching, hotword customization, and satisfying other real-world application requirements. Experimental results show that while most LLM-based ASR systems achieve strong performance on open-source benchmarks, they often underperform on real industry evaluation sets. Thanks to production-oriented optimizations, Fun-ASR achieves state-of-the-art performance on real application datasets, demonstrating its effectiveness and robustness in practical settings. The code and models are accessible at https://github.com/FunAudioLLM/Fun-ASR .
Source: arXiv (Cornell University)
Authors: An, Keyu, Chen, Yanni, Chen, Zhigao, Deng, Chong, Du, Zhihao, Gao, Changfeng, Gao, Zhifu, Gong, Bo, Li, Xiangang, Li, Yabin, Liu, Ying, Lv, Xiang, Ji, Yunjie, Jiang, Yiheng, Ma, Bin, Luo, Haoneng, Ni, Chongjia, Pan, Zexu, Peng, Yiping, Peng, Zhendong, Wang, Peiyao, Wang, Hao, Wang, Haoxu, Wang, Wen, Wang, Wupeng, Wu, Yuzhong, Tian, Biao, Tan, Zhentao, Yang, Nan, Yuan, Bin, Ye, Jieping, Yu, Jixing, Zhang, Qinglin, Zou, Kun, Zhao, Han, Zhao, Shengkui, Zhou, Jingren, Zhu, Yanqiao
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4416292118