HEAR Benchmark
Click to visit website
About
The HEAR benchmark aims to assess audio embedding approaches that generalize across a variety of tasks in everyday domains without fine-tuning. It provides an extensive benchmark suite evaluating audio representations in speech, environmental sounds, and music through 19 diverse tasks. The toolkit includes open-source evaluation code for classification tasks, an API for developing HEAR-compatible models, and a leaderboard to track performance. Launched as part of a NeurIPS competition in 2021, it serves as a foundation for researchers to develop strong audio representations for various applications.
Platform
Task
Features
• open-source evaluation code
• benchmark suite for audio representation
• performance leaderboard
• api for model development
• 19 diverse tasks across multiple audio domains
Average Rating: 0.0
Average Rating: 0.0
5 Stars:
0 Ratings
4 Stars:
0 Ratings
3 Stars:
0 Ratings
2 Stars:
0 Ratings
1 Star:
0 Ratings
User Ratings
No ratings available.
Sign In to Rate this Tool
Alternatives
HEAR Benchmark
A benchmark for evaluating audio representations across diverse tasks in speech, music, and environmental sound.
View DetailsFeatured Tools
Scalytics Connect
A federated AI framework that integrates decentralized data sources for AI development.
View Details