BHASHINI and IISc Develop Sourashtra TTS Model
Ministry of Electronics & IT · Science & Tech · Posted 16 OCT 2026 2:32PM
What happened
The Digital India BHASHINI Division and IISc Bengaluru have developed AI-based Text-to-Speech (TTS) models and a 28+ hour speech corpus for the Sourashtra language to enhance digital access and preservation.
Why it matters for UPSC
This initiative illustrates the application of AI and speech technology for digital inclusion and preservation of low-resource or minority Indian languages, directly relevant to advancements in science and technology and inclusive governance.
Key objectives
- Developed Text-to-Speech (TTS) models for Sourashtra using the VITS architecture.
- Built upon a dedicated Sourashtra speech corpus with over 28 hours of gender-balanced speech recordings created at SPIRE Lab, IISc Bengaluru.
- Made the SPICOR TTS 3.0 Sourashtra Corpus available on BHASHINI’s AIKOSH platform.