Filters








6 Hits in 2.6 sec

M2MeT: The ICASSP 2022 Multi-Channel Multi-Party Meeting Transcription Challenge [article]

Fan Yu, Shiliang Zhang, Yihui Fu, Lei Xie, Siqi Zheng, Zhihao Du, Weilong Huang, Pengcheng Guo, Zhijie Yan, Bin Ma, Xin Xu, Hui Bu
2022 arXiv   pre-print
Along with the dataset, we launch the ICASSP 2022 Multi-channel Multi-party Meeting Transcription Challenge (M2MeT) with two tracks, namely speaker diarization and multi-speaker ASR, aiming to provide  ...  The meeting scenario is one of the most valuable and, at the same time, most challenging scenarios for the deployment of speech technologies.  ...  In order to inspire research on advanced meeting rich transcription, we propose the Multi-channel Multi-party Meeting Transcription Challenge (M2MeT) based on the new AliMeeting corpus introduced above  ... 
arXiv:2110.07393v3 fatcat:cp5m7edtpvgtvcpbywgmh7dje4

Summary On The ICASSP 2022 Multi-Channel Multi-Party Meeting Transcription Grand Challenge [article]

Fan Yu, Shiliang Zhang, Pengcheng Guo, Yihui Fu, Zhihao Du, Siqi Zheng, Weilong Huang, Lei Xie, Zheng-Hua Tan, DeLiang Wang, Yanmin Qian, Kong Aik Lee (+4 others)
2022 arXiv   pre-print
The ICASSP 2022 Multi-channel Multi-party Meeting Transcription Grand Challenge (M2MeT) focuses on one of the most valuable and the most challenging scenarios of speech technologies.  ...  The M2MeT challenge has particularly set up two tracks, speaker diarization (track 1) and multi-speaker automatic speech recognition (ASR) (track 2).  ...  The ICASSP 2022 Multi-Channel Multi-Party Meeting Transcription Challenge (M2MeT) 1 was designed with the aim to provide a common evaluation platform and a sizable dataset for Mandarin meeting transcription  ... 
arXiv:2202.03647v2 fatcat:l3aarvgpxjeoxmtlzzykwy7mie

The Volcspeech system for the ICASSP 2022 multi-channel multi-party meeting transcription challenge [article]

Chen Shen, Yi Liu, Wenzhi Fan, Bin Wang, Shixue Wen, Yao Tian, Jun Zhang, Jingsheng Yang, Zejun Ma
2022 arXiv   pre-print
This paper describes our submission to ICASSP 2022 Multi-channel Multi-party Meeting Transcription (M2MeT) Challenge.  ...  Serialized output training is adopted to multi-speaker overlapped speech recognition. We propose a neural front-end module to model multi-channel audio and train the model end-to-end.  ...  INTRODUCTION Recently, multi-channel multi-party meeting transcription has attracted increasing research interest.  ... 
arXiv:2202.04261v2 fatcat:xjcyz3vofzg4bbfmkddmm26d3i

The CUHK-TENCENT speaker diarization system for the ICASSP 2022 multi-channel multi-party meeting transcription challenge [article]

Naijun Zheng, Na Li, Xixin Wu, Lingwei Meng, Jiawen Kang, Haibin Wu, Chao Weng, Dan Su, Helen Meng
2022 arXiv   pre-print
This paper describes our speaker diarization system submitted to the Multi-channel Multi-party Meeting Transcription (M2MeT) challenge, where Mandarin meeting data were recorded in multi-channel format  ...  In these meeting scenarios, the uncertainty of the speaker number and the high ratio of overlapped speech present great challenges for diarization.  ...  M2MET CHALLENGE DATASET The M2MeT challenge provides a sizeable real-recorded Mandarin meeting corpus called AliMeeting for speaker diarization task (Track1) and ASR task (Track2).  ... 
arXiv:2202.01986v1 fatcat:wwdv4cizlzacfm4hfe5ho63tam

The USTC-Ximalaya system for the ICASSP 2022 multi-channel multi-party meeting transcription (M2MeT) challenge [article]

Maokui He, Xiang Lv, Weilin Zhou, JingJing Yin, Xiaoqi Zhang, Yuxuan Wang, Shutong Niu, Yuhang Cao, Heng Lu, Jun Du, Chin-Hui Lee
2022
Transcription (M2MeT) challenge.  ...  We propose two improvements to target-speaker voice activity detection (TS-VAD), the core component in our proposed speaker diarization system that was submitted to the 2022 Multi-Channel Multi-Party Meeting  ...  Our best result achieved DER 7.80/9.14% on the Eval/Test set. SYSTEM DESCRIPTION Figure 1 illustrates our overall speaker diarization system for the 2022 M2MeT challenge.  ... 
doi:10.48550/arxiv.2202.04855 fatcat:pt3bdr2cdzcohft4xrcdlhsqry

Royalflush Speaker Diarization System for ICASSP 2022 Multi-channel Multi-party Meeting Transcription Challenge [article]

Jingguang Tian, Xinhui Hu, Xinkang Xu
2022 arXiv   pre-print
This paper describes the Royalflush speaker diarization system submitted to the Multi-channel Multi-party Meeting Transcription Challenge(M2MeT).  ...  First, we propose an architecture of combining the multi-channel and U-Net-based models, aiming at utilizing the benefits of these two individual architectures, for far-field overlapped speech detection  ...  Multi-channel Multi-party Meeting Transcription(M2MeT) Challenge [8] is a newly launched ICASSP2022 Signal Processing Grand Challenge.  ... 
arXiv:2202.04814v2 fatcat:rsbvopptbvh5la4cedbbfjpasq