MagicData

sign in

ISCSLP 2022

Conversational Short-phrase Speaker Diarization Challenge

Schedule

Date
Agenda
2022-07-22
Registration Deadline
2022-07-24
Open Training Set and Evaluation Metrics
2022-09-13
Open Evaluation Set
2022-09-15
Final Submission Deadline
2022-09-16
Announcement of Results and Rankings

* Attention: All times above are in (GMT+8)

Committee / Support Team

Members
Affiliation
Yan Yonghong
Chief Scientist and Professor, Institute of Acoustics, Chinese Academy of Sciences, China
Cheng Gaofeng
Assistant Professor, Institute of Acoustics, Chinese Academy of Sciences, China
Xie Lei
Professor, School of Computer Science, Northwestern Polytechnical University, China
Kong Aik Lee
Senior Scientist, A*STAR Institute of Information and Communication, Singapore
Zhang Qingqing
Founder and CEO of Magic Data, China
Qian Yanmin
Associate Professor, School of Artificial Intelligence, Shanghai Jiao Tong University, China
Zhang Pengyuan
Professor, Institute of Acoustics, Chinese Academy of Sciences, China

Competition Result

Rank
TEAM
CDER
1
xlteam (SJTU) - Tao Liu
0.080
2
voicecomm (voicecomm) - kai
0.081
3
TSUP (NPU & TRANSSION) - BoWen Pang
0.091
4
AINJ (亚信科技) - lxf
0.099
5
Zju00 (Zhejiang University) - Jun Wang
0.109
6
SSP (SSP-lab) - Xie Huo
0.285

Winners' Final Submissions (score scaled)

Articles

Find Datasets in Magichub

Quick Recommanded

Download Rank

ASR-RAMC-BigCCSC: A Chinese Conversational Speech Corpus
ASR-SCKwsptSC: A Scripted Chinese Keyword Spotting Speech Corpus
ASR-CCSC: A Chinese Conversational Speech Corpus
ASR-IndoCSC: An Indonesian Conversational Speech Corpus
ASR-CCantCSC: A Chinese Cantonese (Canton) Conversational Speech Corpus
ASR-CStrMAcstCSC: A Chinese Strong Mandarin Accent Conversational Speech Corpus
ASR-MultiDeviCCSC: A Chinese Conversational Speech Corpus
ASR-SCKwsptSC: A Scripted Chinese Keyword-spotting Speech Corpus
ASR-CCMeetingSC: A Chinese Conversational Meeting (Web) Speech Corpus

Recent

NLP_Chinese-Japanese Parallel Corpus - E-Commerce
TTS-SCFChilSC: A Scripted Chinese Female Child's Speech Corpus
ASR-MultiDeviCCSC: A Chinese Conversational Speech Corpus
ASR-SFDuSC: A Scripted Filipino Daily-use Speech Corpus
ASR-CTeleCSC: A Chinese Telephone Conversational Speech Corpus
ASR-CCMeetingSC: A Chinese Conversational Meeting (Web) Speech Corpus
ASR-SCKwsptSC: A Scripted Chinese Keyword-spotting Speech Corpus
ASR-CabNois: A Cabin Noise Dataset

Datasets Customized Service

Magichub is an open source datasets community, while Magic Data still running commercial dataset producing business, including Data Gathering, Labeling, and Annotation Platform Deployment.

Free datasets in Magichub is on puepose for basic modeling training / validation / testing. In mean while customizing specific datasets is possible.

Just feel free telling us your demand, and once the message received, we would contact you as soon as possible.