Audio
12d agoOpen ASR Leaderboard Adds First Global South Language
The Open ASR Leaderboard now includes its first language from the Global South. This update expands the benchmark's reach beyond high-resource languages to track speech recognition accuracy in underrepresented regions. Hugging Face aims to incentivize better model performance for diverse dialects.
The Signal
Practitioners can now evaluate audio models against more inclusive, real-world datasets.