> ## Documentation Index
> Fetch the complete documentation index at: https://developer.eka.care/llms.txt
> Use this file to discover all available pages before exploring further.

# Spotlight

> Open models, datasets and research for AI in Indian healthcare — and who is building on them.

Open models, datasets and research for AI in Indian healthcare, built by Eka Care and used by teams around the world.

<CardGroup cols={4}>
  <Card title="Models" icon="microchip" href="/eka-medai/models">
    Parrotlet ASR, vision and embedding models
  </Card>

  <Card title="Datasets" icon="database" href="/eka-medai/datasets">
    Evaluation sets, knowledge graphs, population data
  </Card>

  <Card title="Resources" icon="toolbox" href="/eka-medai/resources">
    KARMA, BODHI, MCP knowledge bases, reference lists
  </Card>

  <Card title="Live demo" icon="play" href="https://medai.eka.care/">
    Try the models in your browser
  </Card>
</CardGroup>

## Latest Releases

<CardGroup cols={3}>
  <Card title="Parrotlet Open Models" icon="file-lines" href="https://info.eka.care/services/releasing-parrotlet-open-models-open-weights-for-medical-document-intelligence">
    Open weights for medical document intelligence: Parrotlet-v 2.5 Pro, Med Doc Classifier and Document PII Redactor.
  </Card>

  <Card title="Parrotlet-a 2.5 Pro" icon="microphone" href="https://info.eka.care/services/benchmark-update-releasing-parrotlet-a-2-5-pro">
    Medical ASR for Indian English, Hindi, Marathi, Kannada and Telugu, benchmarked against nine production systems.
  </Card>

  <Card title="BODHI" icon="diagram-project" href="https://info.eka.care/services/bodhi-bharat-ontology-for-disease-healthcare-informatics">
    Bharat Ontology for Disease & Healthcare Informatics: open, clinician-validated clinical knowledge graphs.
  </Card>
</CardGroup>

## Blogs

* [Releasing Parrotlet Open Models: Open Weights for Medical Document Intelligence](https://info.eka.care/services/releasing-parrotlet-open-models-open-weights-for-medical-document-intelligence) <span className="eka-tags"><span className="eka-tag">Vision</span><span className="eka-tag">LLM</span><span className="eka-tag">Open Weights</span></span>
* [Benchmark Update: Releasing Parrotlet-a 2.5 Pro](https://info.eka.care/services/benchmark-update-releasing-parrotlet-a-2-5-pro) <span className="eka-tags"><span className="eka-tag">ASR</span><span className="eka-tag">LLM</span><span className="eka-tag">Evaluation</span></span>
* [Beyond Frontier Models: Grounding AI in Indian Clinical Reality](https://info.eka.care/services/beyond-frontier-models-grounding-ai-in-indian-clinical-reality) <span className="eka-tags"><span className="eka-tag">LLM</span><span className="eka-tag">MCP</span><span className="eka-tag">Evaluation</span></span>
* [BODHI: Bharat Ontology for Disease & Healthcare Informatics](https://info.eka.care/services/bodhi-bharat-ontology-for-disease-healthcare-informatics) <span className="eka-tags"><span className="eka-tag">Knowledge Graph</span><span className="eka-tag">Clinical</span><span className="eka-tag">Open Source</span></span>
* [Parrotlet-a 2 Pro: The Most Performant Speech Recognition Model for Indian Healthcare](https://info.eka.care/services/releasing-parrotlet-a-v2-the-gold-standard-for-medical-speech-recognition-in-indian-healthcare) <span className="eka-tags"><span className="eka-tag">ASR</span><span className="eka-tag">LLM</span><span className="eka-tag">Evaluation</span></span>
* [Introducing KARMA – OpenMedEvalKit](https://info.eka.care/services/introducing-karma-openmedevalkit-an-open-source-framework-for-medical-ai-evaluation) <span className="eka-tags"><span className="eka-tag">Evaluation</span><span className="eka-tag">Open Source</span></span>
* [Parrotlet-E & Eka IndicMTEB](https://info.eka.care/services/introducing-parrotlet-e-and-eka-indicmteb-bridging-indias-multilingual-healthcare-gap) <span className="eka-tags"><span className="eka-tag">Embeddings</span><span className="eka-tag">Multilingual</span><span className="eka-tag">Benchmark</span></span>
* [Beyond Traditional WER: Semantic WER for ASR](https://info.eka.care/services/beyond-traditional-wer-the-critical-need-for-semantic-wer-in-asr-for-indian-healthcare) <span className="eka-tags"><span className="eka-tag">ASR</span><span className="eka-tag">Evaluation</span></span>
* [Parrotlet-V Lite (4B) – Vision LLM](https://info.eka.care/services/parrotlet-v-lite-4b-releasing-our-purpose-built-vision-llm-for-parsing-medical-records) <span className="eka-tags"><span className="eka-tag">Vision</span><span className="eka-tag">LLM</span><span className="eka-tag">Open Weights</span></span>
* [Parrotlet-A EN (5B) – Medical ASR LLM](https://info.eka.care/services/parrotlet-a-en-5b-releasing-our-purpose-built-llm-for-english-asr-in-indian-healthcare) <span className="eka-tags"><span className="eka-tag">ASR</span><span className="eka-tag">LLM</span><span className="eka-tag">Open Weights</span></span>
* [Advancing Healthcare AI Evaluation in India](https://info.eka.care/services/advancing-healthcare-ai-evaluation-in-india-ekacare-releases-four-evaluation-datasets) <span className="eka-tags"><span className="eka-tag">Datasets</span><span className="eka-tag">Evaluation</span></span>
* [Lab-Ready and Prescription-Perfect: Eka Care's Small LLMs vs. Industry Giants](https://info.eka.care/services/lab-ready-and-prescription-perfect-ekas-small-llms-vs-industry-giants) <span className="eka-tags"><span className="eka-tag">Vision</span><span className="eka-tag">LLM</span><span className="eka-tag">Evaluation</span></span>
* [Extracting Structured Information from Lab Reports: Challenges and Learnings](https://info.eka.care/services/eka-smart-report) <span className="eka-tags"><span className="eka-tag">Documents</span><span className="eka-tag">Lab Reports</span></span>
* [Eka's heart rate monitor](https://www.eka.care/services/get-the-pulse) <span className="eka-tags"><span className="eka-tag">PPG</span><span className="eka-tag">Signal Processing</span></span>
* [Leveraging semantic technologies in healthcare](https://info.eka.care/services/leveraging-ontologies-in-health-care) <span className="eka-tags"><span className="eka-tag">Ontology</span><span className="eka-tag">Knowledge Graph</span></span>
* [The Quiet Threat: Understanding Latent Iron Deficiency](https://www.eka.care/services/the-quiet-threat-understanding-latent-iron-deficiency) <span className="eka-tags"><span className="eka-tag">Population Health</span><span className="eka-tag">Lab Data</span></span>
* [Cholesterol Chronicles – The Indian Disadvantage](https://info.eka.care/services/cholesterol-chronicles-the-indian-disadvantage-2) <span className="eka-tags"><span className="eka-tag">Population Health</span><span className="eka-tag">Lab Data</span></span>
* [Annual Health Report – India, 2022](https://c.eka.care/insights_2022.pdf) <span className="eka-tags"><span className="eka-tag">Population Health</span><span className="eka-tag">Report</span></span>

## Used by the Community

External teams building on or citing Eka's open datasets, models and research.

<CardGroup cols={4}>
  <Card title="Google Health" href="https://github.com/Google-Health/medasr/blob/main/notebooks/fine_tune_with_hugging_face.ipynb">
    MedASR fine-tuning tutorial uses the Eka Medical ASR dataset
  </Card>

  <Card title="ElevenLabs" href="https://elevenlabs.io/blog/scribe-v2-medical-generally-available">
    Scribe v2 Medical benchmarked on the Eka Medical ASR dataset
  </Card>

  <Card title="Microsoft Research India" href="https://arxiv.org/abs/2609.17355">
    Two papers with Ashoka University analyse Eka's clinical conversation and ASR datasets
  </Card>

  <Card title="Academic research" href="https://aclanthology.org/2026.eacl-industry.40.pdf">
    EACL 2026, Univ. of Hamburg, NORDIS TECH and others fine-tune or benchmark on Eka assets
  </Card>
</CardGroup>

<Accordion title="All citations">
  **Building on Eka open assets**

  * **Google Health** – [MedASR fine-tuning notebook](https://github.com/Google-Health/medasr/blob/main/notebooks/fine_tune_with_hugging_face.ipynb) · fine-tunes on the [Eka Medical ASR dataset](https://huggingface.co/datasets/ekacare/eka-medical-asr-evaluation-dataset)
  * **ElevenLabs** – [Scribe v2 Medical GA](https://elevenlabs.io/blog/scribe-v2-medical-generally-available) · benchmarks on the [Eka Medical ASR dataset](https://huggingface.co/datasets/ekacare/eka-medical-asr-evaluation-dataset)
  * **USTC / Univ. of Florida** – [Synthetic Doctor-Patient Dialogue Generation for Robust Medical ASR](https://aclanthology.org/2026.eacl-industry.40.pdf) (EACL 2026) · fine-tunes [Parrotlet-a-en-5b](https://huggingface.co/ekacare/parrotlet-a-en-5b)
  * **NORDIS TECH** – [MedWER](https://arxiv.org/abs/2609.05728) · baselines on the [Eka Medical ASR dataset](https://huggingface.co/datasets/ekacare/eka-medical-asr-evaluation-dataset)
  * **Univ. of Hamburg** – [MoDiCoL](https://arxiv.org/abs/2606.14459) · sources medical speech from the [Eka Medical ASR dataset](https://huggingface.co/datasets/ekacare/eka-medical-asr-evaluation-dataset)

  **Research citing Eka**

  * **Microsoft Research India / Ashoka** – Jaiswal et al. (2026), [Evaluating Ambient Clinical Scribes in India](https://arxiv.org/abs/2609.17355) · [Eka Clinical Note Generation](https://huggingface.co/datasets/ekacare/clinical_note_generation_dataset) and Medical ASR datasets
  * **Microsoft Research India / Ashoka / IIT Madras** – Shailya et al. (2026), [Building a Cultural Perspective on Doctor-Patient Conversations](https://arxiv.org/abs/2609.18390) · [Eka Clinical Note Generation dataset](https://huggingface.co/datasets/ekacare/clinical_note_generation_dataset)
  * **Kumar et al. (2025)** – [ASR Under the Stethoscope](https://arxiv.org/abs/2512.10967) · [Eka Medical ASR dataset](https://huggingface.co/datasets/ekacare/eka-medical-asr-evaluation-dataset)
</Accordion>

## Publications

* **When De-noising Hurts: A Systematic Study of Speech Enhancement Effects on Modern Medical ASR Systems** · Chondhekar, Murukuri, Vasani, Goyal, Badami, Rana, SN, Pandia, Katiyar, Jagadeesh, Gulati (2025) · [arXiv](https://arxiv.org/abs/2512.17562)
* **Latent iron deficiency in Indian women and children: A descriptive analysis** · Badami, SN, Vasani, Tuniki, Kapoor, Gulati (2024) · *International Journal of Clinical Biochemistry and Research* · [Article](https://www.ijcbr.in/article-details/22074) · DOI 10.18231/j.ijcbr.2024.010
* **Smartphone-based health monitoring in India: Data collection and evaluation for pulse rate estimation** · Shetty, Narasimhamurthy, Nataraj, Prabhu, Jagadeesh, Katre et al. (2024) · *Journal of Family Medicine and Primary Care*
* **AI Horizons in Indian Healthcare: A Vision for Transformation and Equity** · Kapoor, SN, PS, Sethia, DB · *Indian Journal of Community Medicine* · in print
* **Demographic Profile and Antibiotic Prescription Practices in Outpatient Private Practice for Respiratory Tract Infections in Urban India** · Arunkumar, SN, Kapoor, Gupta, Jainkuniya, Gulati · *Cureus* · under review


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.