BHASHINI and IISc Develop Text-to-Speech Models for Sourashtra
Digital India BHASHINI has collaborated with IISc Bengaluru to develop advanced Text-to-Speech models for the Sourashtra language. The technology converts written Sourashtra text directly into synthesized spoken audio using scripts in both Sourashtra and Tamil.
What happened
Digital India BHASHINI collaborated with the Indian Institute of Science (IISc), Bengaluru, to build Text-to-Speech (TTS) tools for the Sourashtra language. Announced on October 16, 2026, the project successfully converted written Sourashtra text directly into synthesized spoken audio.
The speech models and corpus were developed at the SPIRE Lab at IISc Bengaluru. A total of four AI models were created, offering male and female voice outputs across two distinct writing scripts: the Sourashtra script and the Tamil script. The models utilize the VITS deep learning architecture to generate speech directly from character-level text without requiring a separate phonemizer.
The models were trained on the SPICOR TTS 3.0 Sourashtra Corpus, which features more than 28 hours of studio-recorded, gender-balanced audio comprising over 21,400 clean utterances. This dataset has been released on BHASHINI's AIKOSH platform, serving as an open-access repository of language datasets and digital resources for Indian language artificial intelligence.
Important dates
| Announcement Date | 16 October 2026 |
|---|
📌 Static GK links
- Lead Institution: IISc Bengaluru (founded in 1909)
- Language Category: Indo-Aryan language
- Primary Demography: Sourashtra community in Tamil Nadu
- Parent Administrative Body: Digital India BHASHINI Division under MeitY, Government of India
What candidates should do
- Note down the collaborating organizations, the target language, and its primary demographic region for upcoming banking and civil services exams.
- Revise key technological terms such as the VITS architecture and the AIKOSH platform used in language AI projects.
- Practice related multiple-choice questions focusing on digital public infrastructure initiatives under MeitY.
PrepZen analysis
This development marks a significant step forward in preserving and promoting minority and endangered Indian languages through digital technology. By leveraging advanced deep learning frameworks like VITS and releasing open-access resources via the AIKOSH platform, the initiative bridges the digital divide for communities with fewer online language resources. For aspirants preparing for UPSC, State PSC, and Banking examinations, questions from this topic are likely to target institutional collaborations, technological architectures, and linguistic classifications.
Practice questions
Q1. Which institutions developed the Text-to-Speech models for the Sourashtra language announced in October 2026?
Q2. Sourashtra is primarily classified under which language family and spoken predominantly in which state?
Q3. What is the name of BHASHINI's open-access repository where the SPICOR TTS 3.0 Sourashtra Corpus has been released?
Technology converts written Sourashtra text directly into synthesized spoken audio.
Sourashtra is an Indo-Aryan language spoken predominantly within Tamil Nadu.
🎯 Key Point: IISc and BHASHINI build speech synthesis models for Sourashtra language.
FAQs
What is the primary function of the newly developed Sourashtra TTS models?
The models convert written Sourashtra text directly into synthesized spoken audio using male and female voice outputs across Sourashtra and Tamil writing scripts.
Which laboratory at IISc Bengaluru developed the speech corpus and models?
The models were developed at the SPIRE Lab (Signal Processing Interpretation and REpresentation Lab / Speech Processing and Intelligent Research in Engineering Lab) at IISc Bengaluru.
What deep learning architecture do the Sourashtra TTS models use?
The models utilize the VITS (Variational Inference with adversarial learning for end-to-end Text-to-Speech) architecture to generate speech directly from character-level text.
How many hours of audio data are included in the SPICOR TTS 3.0 Sourashtra Corpus?
The dataset features more than 28 hours of studio-recorded, gender-balanced audio comprising over 21,400 clean utterances.
Under which ministry does the Digital India BHASHINI Division function?
The Digital India BHASHINI Division functions under the Digital India Corporation, Ministry of Electronics and Information Technology (MeitY), Government of India.
Sources and references
- anantamias.com (anantamias.com)
- gtmn.in (gtmn.in)
- iisc.ac.in (syspin.iisc.ac.in)
- keekli.in (keekli.in)
Prepared by PrepZen for exam preparation. Consult the linked sources for original announcements and current details. Corrections policy. Spotted an error? Write to prepzen.app@gmail.com.
Practice Quiz — 17 October 2026
Q1. Regarding the appointment of Dr. Sudha Yadav as Governor of Karnataka, consider the following statements: 1. She is appointed by the President under Article 155 of the Constitution. 2. She is the first-ever woman Governor of Karnataka. 3. She replaced Thaawarchand Gehlot. Which of the statements given above are correct?
Q2. Bathukamma, the famous nine-day floral festival of Telangana, begins with 'Engilipula Bathukamma' and concludes with which of the following?
Q3. Which newly designated district was carved out of Kargil district to become the 8th district of the Union Territory of Ladakh?
Q4. Under which statutory act did the Reserve Bank of India sanction the Scheme of Amalgamation between Emirates NBD's Indian branches and RBL Bank?
Q5. What percentage of stake in REC Limited had Power Finance Corporation Limited (PFC) acquired in 2019 prior to the proposed merger?
Keep Learning with PrepZen
Get daily current affairs, live mock tests, mistake bank, vocab builder & more — all free on your phone.