IIT Madras-incubated Bodhan AI has launched four open AI models covering speech, translation and OCR to support multilingual education across Indian languages.
India is building multilingual, voice-first AI through platforms such as BHASHINI, BharatGen and Bharat-VISTAAR to broaden access beyond English-first users.
Gnani.ai has launched Prisma v2.5, a speech-to-text model trained on Indic speech data to improve voice AI accuracy for Indian enterprises.
Meta enhances its AI glasses with Spotify integration, Indian language support including Kannada and Telugu, and improved noise filtering for voice interactions.
OpenAI introduces IndQA, a benchmark spanning 12 Indian languages and 10 cultural domains to test models on reasoning and regional context.
Sarvam AI launches Samvaad, a conversational AI platform that enables enterprises to deploy multilingual voice agents across 11 Indic languages.
