Gnani AI Launches Prisma v3.0, a Speech-to-Text Model Built for India’s Code-Mixed Phone Calls

Bengaluru-based Gnani AI has launched Prisma v3.0, a speech-to-text model trained on Indian telephone audio and code-mixed speech, covering nine Indian languages plus Indian English with four transcription modes and a lower word error rate than ElevenLabs Scribe v2 in its Vistaar evaluation.