no code implementations • 17 Oct 2022 • Adnan Haider, Tim Ng, Zhen Huang, Xingyu Na, Antti Veikko Rosti
Maximum mutual information (MMI) has become one of the two de facto methods for sequence-level training of speech recognition acoustic models.
no code implementations • 31 Aug 2018 • Jiayu Du, Xingyu Na, Xuechen Liu, Hui Bu
For research community, we hope that AISHELL-2 corpus can be a solid resource for topics like transfer learning and robust ASR.
2 code implementations • 16 Sep 2017 • Hui Bu, Jiayu Du, Xingyu Na, Bengu Wu, Hao Zheng
An open-source Mandarin speech corpus called AISHELL-1 is released.
no code implementations • INTERSPEECH 2016 2016 • Daniel Povey, Vijayaditya Peddinti, Daniel Galvez, Pegah Ghahrmani, Vimal Manohar, Xingyu Na, Yiming Wang, Sanjeev Khudanpur
Models trained with LFMMI provide a relative word error rate reduction of ∼11. 5%, over those trained with cross-entropy objective function, and ∼8%, over those trained with cross-entropy and sMBR objective functions.
Ranked #4 on Speech Recognition on WSJ eval92