About Shunya Labs

We’re a research-forward organization building custom voice models for regional languages and mixed speech.Our goal is to make the voice interface of technology accessible to everyone.

We started our journey with Zero STT, the world’s most accurate transcription model, and expanded intocomplete voice agents that can be deployed in air-gapped environments.

What Sets Us Apart

RESEARCH

First-principles research

Proprietary training methodology, model architecture, and training data to improve baseline accuracy.

MODELS

Built for you

Custom SLMs trained on your corpus, your speakers, your terminology.

SECURITY

Privacy-first solutioning

On-prem deployment and custom agent architectures for enterprise-grade security.

DEPLOYMENT

CPU-first architecture

Lightweight models designed to run on CPUs for maximum accessibility.

PLATFORM

Full stack orchestration

End to end platform for agent orchestration, with custom logic for enterprise workflows.

COMMUNITY

Open-source community

Commitment to open models for the community on Hugging Face.

Meet the Team

Ritu Mehrotra

Ritu Mehrotra

Co-founder & CEO

Ritu Mehrotra
Co-founder & CEO

2X founder. Booking.com, Zomato, Mahindra. Understands how AI fails in regulated environments, and how to make it work. MBA, IIFT Delhi.

Sourav Bandyopadhyay

Sourav Bandyopadhyay

Co-Founder & Chief Scientist

Sourav Bandyopadhyay
Co-Founder & Chief Scientist

PhD Scholar, IIT Kharagpur. Proven LLMs will always hallucinate (using Gödel's Incompleteness Theorem). 12 patents, 18 papers, 23 world records. Obsessed with correctness.

Abhishek Sharma

Abhishek Sharma

Co-founder & CBO

Abhishek Sharma
Co-founder & CBO

Helped build TenMarks (acquired by Amazon) and Dineout (acquired by Swiggy) alongside incredible teams. Learned more from the journey than the outcomes. ISB’Hyderabad (PGPMAX).

Arti Khanijo

Arti Khanijo

Chief Product Officer

Arti Khanijo
Chief Product Officer

Worked with global brands like Airtel, ibibo, and Tencent with over 20 years of experience across wellness and consumer tech-known for building scalable, user-first products that create real human impact at the intersection of technology and purpose.

Built by a Team of Iconoclastic Researchers and Engineers

Research Team

12

patents filed

23

world records

18+

peer-reviewed publications

216+

languages supported

Zero STT:

3.10% word error rate

best in the industry.

Read the papers, benchmarks and model cards behind every number.

Join Us

We're building voice infrastructure for the next decade.

Hiring engineers, researchers and linguists who care about deployable intelligence.