### Speech to Text API - Streams Endpoint

Source: https://www.corti.ai/llms.txt

This endpoint facilitates real-time speech-to-text dictation via WebSocket. It can return interim results at a faster rate than final transcripts.

```APIDOC
## WebSocket /streams

### Description
Provides real-time speech-to-text dictation through a WebSocket connection. Interim results can be configured to be returned more frequently than final transcripts.

### Method
WebSocket

### Endpoint
`/streams`

### Parameters
#### Query Parameters
- **interimResults** (boolean) - Configurable - Determines if interim results are returned at a faster rate.
```

--------------------------------

### Generate Clinical Document

Source: https://www.corti.ai/llms.txt

This endpoint accepts clinical context and generation parameters to produce structured medical documentation. It supports various template keys for different document types and allows specifying the output language and model version.

```APIDOC
## POST /documents

### Description
Transforms transcripts, structured facts, or raw clinical data into any type of medical documentation. Output is structured, EHR-compatible, and specialty-specific.

### Method
POST

### Endpoint
`/documents`

### Parameters
#### Request Body
- **context** (array) - Required - An array of context objects. Each object can have `text` (string), `group` (string), and `source` (string). The type of context can be `facts` (structured data) or transcript text.
- **templateKey** (string) - Required - Identifies the desired output format (e.g., SOAP note, referral letter, discharge summary).
- **outputLanguage** (string) - Required - The target language for the generated document.
- **modelName** (string) - Optional - Selects the model version. Supports pinning to a specific version or using the latest.

### Request Example
```json
{
  "context": [
    {
      "text": "Patient presented with chest pain.",
      "group": "symptoms",
      "source": "transcript"
    }
  ],
  "templateKey": "SOAP_NOTE",
  "outputLanguage": "en-US",
  "modelName": "latest"
}
```

### Response
#### Success Response (200)
- **document** (string) - The generated clinical document in a structured format.
- **structured_output** (object) - Machine-readable structured data alongside human-readable text.

#### Response Example
```json
{
  "document": "## Subjective\nPatient presented with chest pain.\n## Objective\n...",
  "structured_output": {
    "sections": {
      "subjective": {
        "symptoms": [
          {
            "text": "chest pain",
            "source": "transcript"
          }
        ]
      }
    }
  }
}
```
```

--------------------------------

### Speech to Text API - Transcripts Endpoint

Source: https://www.corti.ai/llms.txt

This endpoint is used for file upload and batch audio processing for speech-to-text transcription. It supports various parameters to optimize transcription accuracy for clinical conversations.

```APIDOC
## POST /transcripts

### Description
Processes audio files for medical-grade speech-to-text transcription. This endpoint is suitable for batch processing of audio data.

### Method
POST

### Endpoint
`https://api.{environment}.corti.app/v2/interactions/{id}/transcripts/`

### Parameters
#### Query Parameters
- **primaryLanguage** (string) - Required - The language code for the audio (e.g., `en`, `en-GB`, `de`, `gsw-CH`).
- **isDictation** (boolean) - Optional - Optimizes model behavior for spoken dictation.
- **isMultichannel** (boolean) - Optional - Enables separate transcription for each audio channel.
- **diarize** (boolean) - Optional - Enables speaker separation and identification.
- **participants** (array of objects) - Optional - Assigns channel numbers to clinical roles (e.g., `doctor`, `patient`).
- **modelName** (string) - Optional - Selects the model tier (`base` or latest version).

### Request Body
(Not specified in source, assumed to be the audio file for upload)

### Response
#### Success Response (200)
(Response structure not specified in source)

#### Response Example
(No example provided in source)
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.