Gnani.ai Launches Prisma v2.5 Sovereign Voice AI Model

Gnani.ai launched Prisma v2.5, a speech-to-text model trained on 14 million hours of Indic speech data across 12 languages. The model targets higher accuracy and lower latency for real-world Indian voice environments, including noisy telephony and code-switching. It competes with global firms like ElevenLabs and Wispr Flow.