ConferencingSpeech 2022 Challenge: Non-intrusive Objective Speech Quality Assessment (NISQA) Challenge for Online Conferencing Applications

Yi，Gaoxiong1; Xiao，Wei1; Xiao，Yiming1; Naderi，Babak2; Möller，Sebastian2; Wardah，Wafaa2; Mittag，Gabriel3; Cutler，Ross3; Zhang，Zhuohuang4; Williamson，Donald S.4; Chen，Fei5; Yang，Fuzheng6; Shang，Shidong1

10.21437/Interspeech.2022-10597

2022

Interspeech Conference

2308-457X

1990-9772

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

2022-September

3308-3312

SEP 18-22, 2022

null,Incheon,SOUTH KOREA

C/O EMMANUELLE FOXONET, 4 RUE DES FAUVETTES, LIEU DIT LOUS TOURILS, BAIXAS, F-66390, FRANCE

ISCA-INT SPEECH COMMUNICATION ASSOC

With the advances in speech communication systems such as online conferencing applications, we can seamlessly work with people regardless of where they are. However, during online meetings, speech quality can be significantly affected by background noise, reverberation, packet loss, and network jitter, to name a few. Because of its nature, speech quality is traditionally assessed in subjective tests in laboratories and lately also through crowdsourcing following the international standards from the ITU-T Rec. P.800 series. However, those approaches are costly and cannot be applied to customer data. Therefore, an effective objective assessment approach is needed to evaluate or monitor the speech quality of the ongoing conversation. The ConferencingSpeech 2022 challenge targets the non-intrusive deep neural network models for the speech quality assessment task. We open-sourced a training corpus with more than 86K speech clips in different languages, with a wide range of synthesized and live degradations and their corresponding subjective quality scores through crowdsourcing. 18 teams submitted their models for evaluation in this challenge. The blind test sets included about 4300 clips from wide ranges of degradations. This paper describes the challenge, the datasets, and the evaluation methods and reports the final results.

deep learning non-intrusive model speech quality

Others

English

CPCI-S

Deutsche Stiftung für Herzforschung[MO 1038/32-1];

Acoustics ; Audiology & Speech-Language Pathology ; Computer Science ; Engineering

Acoustics ; Audiology & Speech-Language Pathology ; Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic

WOS:000900724503095

2-s2.0-85140094917

Scopus

1.Tencent Ethereal Audio Lab,China
2.Technical University of Berlin,Germany
3.Microsoft Corp.,United States
4.Indiana University,Bloomington,United States
5.Southern University of Science and Technology,China
6.XiDian University,China

Yi，Gaoxiong,Xiao，Wei,Xiao，Yiming,et al. ConferencingSpeech 2022 Challenge: Non-intrusive Objective Speech Quality Assessment (NISQA) Challenge for Online Conferencing Applications[C]. C/O EMMANUELLE FOXONET, 4 RUE DES FAUVETTES, LIEU DIT LOUS TOURILS, BAIXAS, F-66390, FRANCE:ISCA-INT SPEECH COMMUNICATION ASSOC,2022:3308-3312.