ML Scientist

Connecting Scholars with the Latest Academic News and Career Paths

News

Join the ASVspoof 5 Challenge: Robust Speech Deepfake Detection and Automatic Speaker Verification – Registration Open Now

Registration for the ASVspoof 5 Challenge is now open! The challenge focuses on two tracks: Speech deepfake detection (DF) and Spoofing-robust automatic speaker verification (SASV). ASVspoof is a community-driven, not-for-profit challenge series that aims to develop and benchmark generalizable speech deepfake detection and automatic speaker verification systems. The challenge data is constructed using public speech resources, and organizers provide baseline systems and reference metrics. This edition of the challenge involves a larger amount of data, focusing on non-studio-quality speech data to promote robustness and practical applications.

To participate, follow these steps:

The challenge timeline is as follows:

  • Training and development data available: May 20, 2024
  • Challenge leaderboard (Codalab) opens: June 05, 2024
  • Evaluation data available: June 12, 2024
  • Challenge submissions due: July 17, 2024
  • ASVspoof 5 workshop paper deadline: July 31, 2024
  • Acceptance notifications: August 10, 2024
  • ASVspoof 5 workshop at Interspeech: August 31, 2024

Access to Codalab will be granted to registered participants only. For more information, refer to the evaluation plan, which will be updated as the challenge progresses.

Don't miss this opportunity to contribute to the field of speech deepfake detection and automatic speaker verification!

Leave a Reply

Your email address will not be published. Required fields are marked *