Sarvam AI has introduced 'Indic DiarBench,' a new benchmark dataset designed to evaluate speech recognition and speaker attribution across 22 Indian languages.
The project, developed in collaboration with AI4Bharat, allocates two hours of data per language to improve the accuracy of AI-driven transcription tools.
This initiative aims to bridge the gap in linguistic AI development, enabling more robust voice-based applications for diverse Indian user bases.