BHASHINI and IISc Bengaluru Develop AI Text-to-Speech Models for Sourashtra Language

Key facts

• The Digital India BHASHINI Division (DIBD) under the Digital India Corporation (DIC), Ministry of Electronics and Information Technology (MeitY), in collaboration with the Indian Institute of Science (IISc), Bengaluru, has developed Text-to-Speech (TTS) models for Sourashtra, a minority Indo-Aryan language spoken predominantly by the Sourashtra community in Tamil Nadu.
• The models use the VITS (Variational Inference with adversarial learning for end-to-end Text-to-Speech) architecture and generate Sourashtra speech directly from text, with male and female voices each available in Tamil-script and Sourashtra-script versions, making four models in total.
• The models were trained on the SPICOR TTS 3.0 Sourashtra Corpus developed at the SPIRE Lab, IISc Bengaluru, containing over 28 hours of gender-balanced speech recordings, now available on BHASHINI's AIKOSH platform.
• The male voice model was trained on approximately 10,000 utterances and the female voice model on approximately 11,400 utterances; the models generate audio at 22,050 Hz and operate directly on character-level text without a separate phonemizer.
• Sourashtra is spoken by around 2.48 lakh people according to the 2011 Census and has a relatively limited digital presence.
• BHASHINI supports 36 Indian text languages, 23 Indian voice languages and 35 international languages.