CHiME-6 Challenge: Tackling Multispeaker Speech Recognition for Unsegmented Recordings

Задача CHiME-6: решение задачи распознавания речи нескольких дикторов в неразмеченных записях
Emmanuel Vincent A, Daniel Povey, Sanjeev P. Khudanpur, David Snyder, Shinji Watanabe, Jon Barker, Takuya Yoshioka, Vimal Manohar, Naoyuki Kanda, Yusuke Fujita, Shota Horiguchi, Desh Raj, Christoph Boeddeker, Aswin Shanmugam Subramanian, Xuankai Chang, Neville Ryant, Zhaoheng Ni, Jan Trmal, Michael Mandel, Ashish Arora, Bar Ben-Yair
2020-05-04

CHiME-6 Challengemultispeaker speech recognitionspeaker diarizationspeech enhancementunsegmented recordings
Following the success of the 1st, 2nd, 3rd, 4th and 5th CHiME challenges we organize the 6th CHiME Speech Separation and Recognition Challenge (CHiME-6). The new challenge revisits the previous CHiME-5 challenge and further considers the problem of distant multi-microphone conversational speech diarization and recognition in everyday home environments. Speech material is the same as the previous CHiME-5 recordings except for accurate array synchronization. The material was elicited using a dinner party scenario with efforts taken to capture data that is representative of natural conversational speech. This paper provides a baseline description of the CHiME-6 challenge for both segmented multispeaker speech recognition (Track 1) and unsegmented multispeaker speech recognition (Track 2). Of note, Track 2 is the first challenge activity in the community to tackle an unsegmented multispeaker speech recognition scenario with a complete set of reproducible open source baselines providing speech enhancement, speaker diarization, and speech recognition modules.
1
CHiME-6 targets distant multi-microphone conversational speech diarization and recognition in everyday home environments.
2
The challenge uses dinner-party recordings designed to represent natural multispeaker conversational speech, with accurate microphone-array synchronization added to the CHiME-5 material.
3
Track 1 provides a baseline for segmented multispeaker speech recognition.
4
Track 2 addresses unsegmented multispeaker speech recognition and is described as the first community challenge activity for this scenario.
5
Track 2 includes complete reproducible open-source baselines covering speech enhancement, speaker diarization, and speech recognition.

distant multi-microphone conversational speech in everyday home environments, recorded during dinner-party scenarios

multispeaker speech diarization and recognition, including speech separation and enhancement, for segmented and unsegmented recordings

Publication Details
Publication Date
2020-05-04
Journal
Publisher
ISSN
Cited by
377
Access Type
Author Information
Authors
Emmanuel Vincent A
Daniel Povey
Sanjeev P. Khudanpur
David Snyder
Shinji Watanabe
Jon Barker
Takuya Yoshioka
Vimal Manohar
Naoyuki Kanda
Yusuke Fujita
Shota Horiguchi
Desh Raj
Christoph Boeddeker
Aswin Shanmugam Subramanian
Xuankai Chang
Neville Ryant
Zhaoheng Ni
Jan Trmal
Michael Mandel
Ashish Arora
Bar Ben-Yair
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%