{"id":899040,"date":"2022-11-18T12:11:43","date_gmt":"2022-11-18T20:11:43","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-academic-program&p=899040"},"modified":"2023-03-01T01:42:29","modified_gmt":"2023-03-01T09:42:29","slug":"acoustic-echo-cancellation-challenge-icassp-2023","status":"publish","type":"msr-academic-program","link":"https:\/\/www.microsoft.com\/en-us\/research\/academic-program\/acoustic-echo-cancellation-challenge-icassp-2023\/","title":{"rendered":"Acoustic Echo Cancellation Challenge – ICASSP 2023"},"content":{"rendered":"\n\n

<\/p>\n\n\n\n\n\n\n

Program dates:<\/strong> December 2022-February 2023<\/p>\n\n\n\n

The ICASSP 2023 (opens in new tab)<\/span><\/a> Acoustic Echo Cancellation Challenge is intended to stimulate research in the area of\u202facoustic echo cancellation\u202f(AEC), which is an important part of speech enhancement and is still a top issue in audio communication and conferencing systems. This is the fourth AEC challenge. The winners of this challenge are selected based on the average Mean Opinion Score achieved across all different single talk and double talk scenarios, and the speech recognition rate.<\/p>\n\n\n\n

Registration procedure<\/h3>\n\n\n\n

To register for the challenge,\u202fparticipants are required to email Acoustic Echo Cancellation Challenge aec_challenge@microsoft.com (opens in new tab)<\/span><\/a> with the name of their team members, emails, affiliations, team name, and tentative paper title. Participants also need to register on the Challenge CMT (opens in new tab)<\/span><\/a> site where they can submit the enhanced clips.<\/p>\n\n\n\n

Challenge tracks<\/h3>\n\n\n\n

There are two tracks for this challenge:<\/p>\n\n\n\n

    \n
  1. Non-personalized AEC. This is similar to the ICASSP 2022 AEC Challenge.<\/li>\n\n\n\n
  2. Personalized AEC. This adds speaker enrollment for the near end speaker. A speaker enrollment is a 15-25 second recording of the near end speaker that can be used for adopting the AEC for personalized echo cancellation. For training and model evaluation, the datasets in AEC-Challenge Github page (opens in new tab)<\/span><\/a>can be used, which include both echo and near-end only clips from users. For the blind test set, the enrollment clips will be provided.<\/li>\n<\/ol>\n\n\n\n

    Latency and runtime requirements<\/h3>\n\n\n\n

    Algorithmic <\/em><\/strong>laten<\/em><\/strong>cy:<\/em><\/strong> The offset introduced by the whole processing chain including STFT, iSTFT, overlap-add, additional lookahead frames, etc., compared to just passing the signal through without modification. But this doesn\u2019t include buffering latency.<\/p>\n\n\n\n