> ## 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.

# Overview

> EkaScribe turns clinical conversations into structured medical notes — via a simple session-based API

EkaScribe is Eka Care's AI medical scribe. Record or stream a consultation (or a dictation) and get back accurate, structured clinical notes — in the exact format you define — across 10+ Indian and international languages.

## How it works

One session, five calls:

```
Create Session  →  Upload Audio  →  End Session  →  Poll  →  Structured Notes
POST /sessions     POST .../audio    POST .../end     GET /sessions/{id}
```

1. **Create a session** with the note templates you want.
2. **Send audio** — HTTP chunks or a real-time WebSocket stream.
3. **End the session** to start processing.
4. **Get the results** — transcript plus structured notes per template (poll or webhook).

Start with the [Quickstart](https://developer.eka.care/docs/scribe/quickstart.md) — the full flow is \~30 lines of code in any language.

## What you can build with it

> **[Sessions API](https://developer.eka.care/api-sdk/scribe/sessions/overview.md)**
>
> Voice capture and transcription: create sessions, upload or stream audio, retrieve structured results.

> **[Templates API](https://developer.eka.care/api-sdk/scribe/templates/create-template.md)**
>
> Define your own note formats in plain markdown — SOAP notes, specialty OPD notes, discharge summaries.

> **[Documents API](https://developer.eka.care/api-sdk/scribe/documents/create-update-document.md)**
>
> Read, list, edit and manage every note generated on a session.

> **[Try the APIs live](https://developer.eka.care/docs/scribe/authenticate-and-try.md)**
>
> Paste your client credentials and call every EkaScribe API from your browser.

To see the note land in the patient's record with a prescription and a parsed lab report, follow [Scribe with medical records](https://developer.eka.care/docs/solutions/scribe-and-medical-records.md).

## Key capabilities

* **Built on an open protocol** — all APIs implement the [MedScribe Alliance scribe-emr protocol](https://developer.eka.care/docs/scribe/protocol/overview.md); EkaScribe is the first scribe service to open-source it. No vendor lock-in.
* **Custom templates** — describe your note format in markdown; EkaScribe follows it. No rigid schemas.
* **Multilingual** — English, Hindi, Tamil, Telugu, Bengali, Marathi, Gujarati, Kannada, Malayalam, Punjabi. Auto-detection and output translation.
* **Real-time or batch** — WebSocket streaming for live scribing, chunked HTTP upload for recorded audio.
* **Two models** — `pro` (speaker diarization, custom templates, 60-min sessions) and `lite` (fast, dictation-focused).
* **FHIR output** — notes available as a [FHIR bundle](https://developer.eka.care/docs/scribe/resources/fhir-structure.md) for interoperability.
* **Webhooks** — get notified when processing completes instead of polling.

## Integration options

| Option | Best for | Get started |
| - | - | - |
| **REST APIs** | Any language, full control | [Quickstart](https://developer.eka.care/docs/scribe/quickstart.md) |
| **SDKs** (TypeScript, Python, Java, Android, iOS) | Fastest integration — VAD, chunking, retries handled for you | [SDKs](https://developer.eka.care/api-sdk/scribe/sdks/ts-sdk.md) |
| **Chrome Extension** | No-code, works on any web page | [Install](https://chromewebstore.google.com/detail/ekascribe-ai-powered-clin/nncfcjgelepkhpjfkejgkncdfbcfmhom) |

## Integrating with an AI coding assistant?

These docs are LLM-ready. Point Claude, ChatGPT, Cursor or any coding agent at:

* **`https://developer.eka.care/llms.txt`** — index of all pages in plain text
* **`https://developer.eka.care/llms-full.txt`** — the full documentation in one file

Or grab the ready-made **[CLAUDE.md integration file](https://developer.eka.care/docs/scribe/build-with-ai/claude-md.md)** — drop it in your project and your coding agent can build the record → transcribe → note flow in one shot.


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