Dear all,
In order to stimulate the research of non-intrusive objective speech quality assessment for online conferencing applications, we will hold the “Non-intrusive Objective Speech Quality Assessment (NISQA) Challenge
for Online Conferencing Applications (Conferencing Speech 2022)” as a special session of INTERSPEECH 2022. We sincerely invite you to participate in this challenge.
With the advances in speech communication systems such as online conferencing applications, we can seamlessly work with people regardless where they are. However, during the meeting, the speech quality is
significantly affected by background noise, reverberation, packet loss, network jitter and many other factors. Therefore, an effective objective assessment approach is needed to evaluate or monitor the speech quality of the ongoing call. If severely degraded
speech quality is detected, then the statistical information could be shared with the service provider in time. This enables the service provider to find out the root cause and improve the speech quality to guarantee better user experience in the future.
This challenge will provide some speech datasets with subjective ratings which were not published before. These datasets contain hundreds of hours of speech samples with subjective Mean Opinion Score (MOS)
and covering most of the impairment scenarios users might face in on-line speech communication. The final ranking of this challenge will be decided by the accuracy of the predicted MOS scores from the submitted model or algorithm on the test dataset. Each
registered team could participate in the above task. Top two winning teams from the task will be awarded the prizes provided by Tencent Ethereal Audio Lab. The First Prize is 1500 USD. The Second Prize is 800 USD. The internship opportunities at Tencent Ethereal
Audio Lab will be provided to the students with excellent performance. We plan to use the webpage of Tencent Ethereal Audio Lab. (https://tea-lab.qq.com/) as the competition registration entry point, and a baseline system
will be provided by the organizing committee.
A tentative schedule is as follows:
Challenge registration open: January 19, 2022
Release of evaluation plan, the list of training data and development test set: January 27, 2022
Release of baseline system: February 17, 2022
Deadline of challenge registration: March 12, 2022
Release of evaluation test set: March 14, 2022
Deadline of submitting the results: March 17, 2022
Notification of the results of participants: March 20, 2022
Interspeech paper submission deadline: March 21, 2022
For more detail, you can visit the webpage (https://tea-lab.qq.com/conferencingspeech-2022).
If you might be interested in participating in the challenge and writing a paper for the special session, please not reply to this address but ConferencingSpeech@xxxxxxxxxxx.
Regards,
The Conferencing Speech 2022 organizing committee
Gaoxiong Yi, Wei Xiao, Yiming Xiao, Babak Naderi, Sebastian Möller, Gabriel Mittag, Ross Cutler, Zhuohuang Zhang, Donald S. Williamson, Fei Chen, Fuzheng yang and Shidong Shang
--
Donald S. Williamson
Assistant Professor
Department of Computer Science
Luddy School of Informatics, Computing and Engineering
Personal page:
https://homes.luddy.indiana.edu/williads
ASPIRE Research Group:
https://aspire.sice.indiana.edu
https://twitter.com/IUAspireRschGrp
https://www.instagram.com/iuaspirerschgrp/
https://www.youtube.com/channel/UCw6r8rLx5RypKBMeBjEzsjw/