-------------------------------------------------------------------------------------
Pre-announcement
7th CHiME Speech Separation and Recognition Challenge (CHiME-7)
Launch Date: February 1st, 2023
Workshop: Dublin, August 25th, 2023 (TBC)
https://www.chimechallenge.org/---------------------------------------------------------------------------------------
Dear colleague,
It gives us great pleasure to pre-announce the 7th CHiME Speech Separation and Recognition Challenge (CHiME-7)
CHiME-7 TASKS
The CHiME-7 challenge will be composed of two separate tasks:
The first task focuses on distant automatic speech transcription and segmentation with multiple recording devices, especially on generalization across array topologies.
The second task focuses on unsupervised domain adaptation for conversational speech enhancement. It involves learning to suppress noise in multi-speaker conversational speech recordings using out-of-domain labeled data and in-domain unlabeled data.
For further task details, see the CHiME Challenge website,
https://www.chimechallenge.org/WORKSHOP
CHiME-7 Challenge submissions will be presented at a 1-day workshop on August 25th that will be organised as a satellite event of Interspeech 2023 (details TBC). The workshop will also be accepting non-challenge papers on related topics. For further details,
https://www.chimechallenge.org/current/workshop/ FORUM
If you are considering participating, please join the CHiME Google group for discussions and further announcements.
https://groups.google.com/forum/#!forum/chime5/joinIMPORTANT DATES
1 Feb 2023, Challenge opening
30 June 2023, Challenge submission deadline
25 Aug 2023, Interspeech satellite workshop (details TBC)
CHiME STEERING GROUP
Jon Barker, Univ. of Sheffield, UK
Emmanuel Vincent, Inria, France
Shinji Watanabe, Carnegie Mellon Univ., USA
Michael Mandel, CUNY/META, USA
Marc Delcroix, NTT, Japan
Leibny Paola, Garcia Perera Johns Hopkins Univ., USA
TASK 1 ORGANIZERS
Samuele Cornell, Universita Politecnica delle Marche, Italy
Shinji Watanabe, Carnegie Mellon University, USA
Xuankai Chang, Carnegie Mellon University, USA
Yoshiki Masuyama, Tokyo Metropolitan University, Japan
Zhong-Qiu Wang, Carnegie Mellon University, USA
Stefano Squartini, Universita Politecnica delle Marche, Italy
Matthew Wiesner, Johns Hopkins University, USA
Desh Raj, Johns Hopkins University, USA
Paola Garcia, Johns Hopkins University, USA
Sanjeev Khudanpur, Johns Hopkins University, USA
TASK 2 ORGANIZERS
Simon Leglaive, CentraleSupélec and IETR, France
Manuel Pariente, Pulse Audition, France
Mostafa Sadeghi, Inria, France
Efthymios Tzinis, UIUC, USA
Léonie Borne, Pulse Audition, France
Scott Wisdom, Google, USA
John R. Hershey, Google, USA
--
Professor Jon Barker,
Department of Computer Science,
University of Sheffield
+44 (0) 114 222 1824