The IISc SraVaani AI model can recognise speech across 65 Indian languages and dialects, expanding access to voice technology for underserved language communities. Researchers at the Indian Institute of Science (IISc) have released SraVaani, an open-source voice AI model capable of recognising speech in 65 Indian languages and dialects. Developed by IISc’s SPIRE Lab with ARTPARK and support from Google, the model can convert speech into text across 10 scripts and automatically identify the language being spoken. SraVaani covers 20 scheduled languages and 45 regional languages and dialects, including Tulu, Garo, Angika, Chakma and Kokborok. The languages covered are spoken by around 25 crore people.The model is freely available on Hugging Face under an MIT licence, allowing researchers, startups and developers to use and build applications with it. SraVaani was trained using more than 31,000 hours of speech collected from 156,000 people across 165 districts in 28 states through Project Vaani.
IISc SraVaani AI Model Supports 65 Indian Languages
