The VoxCeleb Speaker Recognition Challenge 2020 - Track 4 (Diarisation, open)

Organized by anagrani - Current server time: Nov. 30, 2020, 5:23 p.m. UTC

Previous

Challenge workshop
Sept. 9, 2020, midnight UTC

Current

Permanent
Oct. 16, 2020, 11:59 p.m. UTC

End

Competition Ends
Never

The VoxCeleb Speaker Recognition Challenge (VoxSRC) 2020

Track 4: Diarisation (open)

Welcome to the VoxCeleb speaker recognition challenge! The goal of this challenge is to probe how well current methods can segment and recognize speakers from speech obtained 'in the wild'. The dataset is obtained from YouTube videos, consisting of multi-speaker audio from both professionally edited videos as well as more casual conversational multi-speaker audio in which background noise, laughter, and other artefacts are observed in a range of recording environments.

The task of speaker diarisation is to break up multi-speaker audio into homogenous single speaker segments, effectively solving ‘who spoke when’.

In order to submit a file you need to click on the "Submit/View Results" link under the "Participate" tab. After this, you will be able to see two buttons, corresponding to the Competition Phases. Click on one of the buttons to choose the phase you want to submit to. The available phases are:

  • Challenge workshop: This is the phase during which participants will be submitting their results for the challenge workshop to be held in conjunction with Interspeech 2020. The leaderboard will be visible during this phase.
  • Permanent: During this phase new submissions will be accepted but they will not be taken into account for the Challenge workshop. During this phase the leaderboard will be INVISIBLE until the challenge workshop takes place. It will be made visible after that.

In order to prevent overfitting to the test data participants can only submit one result per day. There is also a limit over the total number of submissions for each phase, see more details about this under "Submit/View Results".

This track follows the Open training data condition, meaning participants can train on any data that they see fit.

Organisers

Arsha Nagrani, VGG, University of Oxford,
Joon Son Chung, Naver, South Korea,
Andrew Zisserman, VGG, University of Oxford,
Jaesung Huh, VGG, University of Oxford,
Ernesto Coto, VGG, University of Oxford,
Andrew Brown, VGG, University of Oxford,
Weidi Xie, VGG, University of Oxford,
Mitchell McLaren, Speech Technology and Research Laboratory, SRI International, CA,
Douglas A Reynolds, Lincoln Laboratory, MIT.

More information

For more information, visit the VoxSRC Challenge page.

Acknowledgements

This work is supported by the EPSRC programme grant Seebibyte EP/M013774/1: Visual Search for the Era of Big Data.

The VoxCeleb Speaker Recognition Challenge (VoxSRC) 2020

Track 4: Diarisation (open)

Evaluation Plan

Please see the validation scoring code and the VoxSRC Challenge page for details on the evaluation.

Terms and Conditions

Participation in this competition is open to all who are interested and willing to comply with the rules laid out under the "Learn the Details" and "Participate" tabs. There is no cost to participate, although entries to the challenge will only be considered if a technical report is submitted on time. This should not affect later publications of your method if you restrict your report to 2 pages including references. You can still submit to the leaderboard, however, even if you do not submit a technical report.

By submitting to the challenge, you allow the organizers to use your submitted scores for any purpose, including analysis and comparison with other teams.

We also kindly ask you to associate the CodaLab account you will use for the competition to your institutional e-mail. We reserve the right to revoke your access to the competition otherwise.

In case of any issues, all decisions made by the Organizing Committee will be final.

You can download a ZIP file with the test data from the VoxConverse website. The list of files that will be used for scoring can be found here. Note that there are 310 files that will be used for scoring. Please prepare a single .rttm file that contains the scores for all 310 audio files for submission to the leaderboard. 

 

For reference, we have added a baseline result to the leaderboard, submitted by the vggoxford user.

Good Luck!

Challenge workshop

Start: Sept. 9, 2020, midnight

Description: Submissions for the challenge workshop that will be held in conjunction with Interspeech 2020

Permanent

Start: Oct. 16, 2020, 11:59 p.m.

Description: Submissions for comparison with previous ones. Not to be taken into account for the challenge workshop

Competition Ends

Never

You must be logged in to participate in competitions.

Sign In