Eka Care, NVIDIA Partner to Develop Offline Multilingual AI Medical Scribe for India

Eka Care is using NVIDIA NeMo Curator to refine and organise datasets of Indian medical conversations to maintain data quality and model reliability.

Eka Care has partnered with NVIDIA to develop a next-generation AI-powered medical scribe, including a variant designed to operate entirely offline, aimed at expanding secure and accessible clinical documentation across India.

The collaboration seeks to build a consolidated, end-to-end multilingual medical scribe model tailored to India’s diverse healthcare landscape.

By enabling offline functionality, the solution is intended to support AI-powered transcription and documentation in remote and resource-constrained clinics where internet connectivity is unreliable.

The partnership will leverage NVIDIA’s AI software stack and compute infrastructure to accelerate development of Eka Care’s proprietary clinical AI systems. By combining Indian clinical datasets with NVIDIA’s AI platforms, the initiative aims to address data privacy, cloud dependency and real-time transcription challenges.

As part of the development process, Eka Care is using NVIDIA NeMo Curator to refine and organise datasets of Indian medical conversations to maintain data quality and model reliability.

The company is also integrating the NVIDIA Nemotron CC dataset during pre-training to prevent catastrophic forgetting, ensuring retention of general language understanding while adapting to medical terminology.

Eka Care is evaluating NVIDIA Nemotron Speech models, including Parakeet automated speech recognition, to power its end-to-end application. The company aims to integrate speech recognition capabilities with large language models into a single architecture designed to reduce computational overhead while maintaining clinical accuracy.

Vikalp Sahni, Founder and CEO of Eka Care, said building a medical scribe for India requires navigating complex code-mixing and diverse accents. He added that the collaboration with NVIDIA will enable development of a “sovereignty-first” model capable of running securely on a doctor’s mobile device without reliance on cloud connectivity.

Tobias Halloran, Director of EMEAI Startups and Venture Capital at NVIDIA, highlighted India’s growing AI startup ecosystem and NVIDIA’s role in supporting founders through accelerated computing, scalable infrastructure and programmes such as NVIDIA Inception and the NVIDIA VC Alliance.

The announcement comes amid momentum around the India AI Impact Summit 2026 hosted by the Government of India under the IndiaAI Mission in New Delhi.

Stay tuned for more such updates on Digital Health News