“OLR Challenge 2017”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
第46行: 第46行:
 
* Publications based on the data provided by the challenge should cite the following paper:  
 
* Publications based on the data provided by the challenge should cite the following paper:  
  
'''Dong Wang, Lantian Li, Difei Tang, Qing Chen, AP16-OL7: a multilingual database for oriental languages and a language recognition baseline, submitted to APSIPA 2016.'''[http://wangd.cslt.org/public/pdf/ole.pdf pdf]
+
'''Dong Wang, Lantian Li, Difei Tang, Qing Chen, AP16-OL7: a multilingual database for oriental languages and a language recognition baseline, APSIPA 2016.'''[http://wangd.cslt.org/public/pdf/ole.pdf pdf]
 +
 
 +
'''Zhiyuan Tang, Dong Wang, Yixiang Chen, Qing Chen: AP17 OLE: Data, plan, and baseline, submitted to APSIPA 2017.'''
  
 
==Important dates==
 
==Important dates==

2017年6月9日 (五) 00:11的版本

Oriental Language Recognition (OLR) 2017 Challenge

Oriental languages involve interesting specialalities. The OLR challenge series aims at boosting language recognition technology for oriental languages. Following the success of OLR Challenge 2016, the new challenge in 2017 follows the same theme, but sets up more challenging tasks in the sense of:

  • more languages: OLR 2016 involves 7 languages, OLR 2017 involves 10 languages.
  • shorter speech segments. OLR 2017 set individual tasks for 1 second, 3 second, 5 second segments separately.

As the first OLR challenge, we will publish the result on a special session of APSIPA 2017. See more details for the AP17 special session.

Data

The challenge is based on two multilingual database, AP16-OL7 that was designed for the OLR challenge 2016, and a new complementary AP17-OL3 database. These two databases are both provided by SpeechOcean (www.speechocean.com).


The features for AP16-OL7 involve:

  • Mobile channel
  • 7 languages in total
  • 24 speakers (18 speakers for training/development, 6 speakers for test).
  • 71 hours of speech signals in total
  • Transcriptions and lexica are provided
  • The data profile is here
  • The Licence for the data is here

The feature for AP17-OL3 involve:

  • Mobile channel
  • 3 languages in total
  • 24 speakers (18 speakers for training/development, 6 speakers for test).
  • 30 hours of speech signals in total
  • Transcriptions and lexica are provided
  • The data profile is here
  • The Licence for the data is here

Evaluation plan

Evaluation tools

  • The Kaldi-based baseline scripts here
  • The evaluation toolkit here

Participation rules

  • Participants from both academy and industry are welcome
  • Publications based on the data provided by the challenge should cite the following paper:

Dong Wang, Lantian Li, Difei Tang, Qing Chen, AP16-OL7: a multilingual database for oriental languages and a language recognition baseline, APSIPA 2016.pdf

Zhiyuan Tang, Dong Wang, Yixiang Chen, Qing Chen: AP17 OLE: Data, plan, and baseline, submitted to APSIPA 2017.

Important dates

  • May 20, AP17-OL7 training data release.
  • Oct.1, test data release.
  • Oct.2, 12:00PM, Beijing time, submission deadline
  • APSIPA17, results announcement.

Registration procedure

If you are interested to participate the challenge, or if you have any other questions, comments, suggestions about the challenge, please send email to the organizer:

  • Dr. Dong Wang (wangdong99@mails.tsinghua.edu.cn)
  • Ms. Qing Chen(chenqing@speechocean.com)


Organizers

  • Dong Wang, Tsinghua University
  • Lantian Li, Tsinghua University
  • Qing Chen, SpeechOcean