HEAR Benchmark

hear-eval-kit

Evaluation kit for the HEAR Benchmark

hear-baseline

Simple baseline model for the HEAR benchmark

hear2021-submitted-models

Open-source audio embedding models, submitted to the HEAR 2021 challenge

hear-preprocess

Dataset preprocessing code for the HEAR 2021 NeurIPS competition

hear-validator

Submission validator for the HEAR Benchmark

hearbenchmark.com

HEAR Benchmark website and leaderboard submissions

hear2021-open-tasks-downsampled

Downsampled versions of the HEAR 2021 open tasks, for rapid testing and CI/CD

.github

hear2021-sample-datasets

HEAR 2021 pre-processed GTZAN Music/Speech Task

torchvggish

Pytorch port of Google Research's VGGish model used for extracting audio features.

hear-leaderboard

Leaderboard scripts for the HEAR benchmark, using TSP and imputation