Leaderboard & Open Tasks & Downstream Evaluations

The leaderboard is now live, you can view it here

The open task datasets and code for running evaluations is now available. Please see the hear-eval-kit for information on downloading the open tasks and how to run evaluation.

If you need any help getting your model running on the open tasks, or on how to use the hear-eval-kit to improve your submission, we would love hear from you.