### Speech Start Event Example

Source: https://docs.gladia.io/api-reference/v2/live/websocket

This example shows the JSON payload for a 'speech_start' event, indicating the beginning of detected speech activity. It includes a session ID, timestamp, event type, and timing data.

```json
{
  "session_id": "550e8400-e29b-41d4-a716-446655440000",
  "created_at": "2025-09-19T12:34:06Z",
  "type": "speech_start",
  "data": {
    "time": 1.24,
    "channel": 0
  }
}
```

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

### Post Chapterization Message Example

Source: https://docs.gladia.io/api-reference/v2/live/websocket

This example shows the structure of a message containing chapterization results for an audio chunk. It includes session ID, creation timestamp, and detailed results with headlines, start/end times, and keywords.

```json
{
  "session_id": "550e8400-e29b-41d4-a716-446655440000",
  "created_at": "2025-09-19T12:46:00Z",
  "type": "post_chapterization",
  "error": null,
  "data": {
    "results": [
      {
        "headline": "Project kickoff",
        "start": 0,
        "end": 60,
        "keywords": [
          "timeline",
          "owner"
        ]
      }
    ]
  }
}
```

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

### Post Summarization Message Example

Source: https://docs.gladia.io/api-reference/v2/live/websocket

This example demonstrates the format of a message containing a post-processing summarization of an audio chunk. It includes session ID, creation timestamp, and the summarized text.

```json
{
  "session_id": "550e8400-e29b-41d4-a716-446655440000",
  "created_at": "2025-09-19T12:47:00Z",
  "type": "post_summarization",
  "error": null,
  "data": {
    "results": "The team aligned on goals and next steps for the quarter."
  }
}
```

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

### Start Recording Event

Source: https://docs.gladia.io/api-reference/v2/live/websocket

This message is emitted when audio capture or streaming begins for a session.

```APIDOC
## Start Recording

### Description
This message contains the informations about the start recording of the audio chunk.

### Method
SEND

### Endpoint
N/A (WebSocket Event)

### Payload Schema
```json
{
  "session_id": "string",
  "created_at": "string",
  "type": "start_recording"
}
```

### Parameters
#### Payload Fields
- **session_id** (string) - Required - UUID identifying the session.
- **created_at** (string) - Required - ISO 8601 timestamp in UTC.
- **type** (string) - Required - Lifecycle event emitted when audio capture/streaming begins. Must be 'start_recording'.

### Request Example
```json
{
  "session_id": "550e8400-e29b-41d4-a716-446655440000",
  "created_at": "2025-09-19T12:34:05Z",
  "type": "start_recording"
}
```
```

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

### Start Session

Source: https://docs.gladia.io/api-reference/v2/live/websocket

Initiates a new audio processing session. This event is emitted once when the session is created and ready.

```APIDOC
## Start Session

### Description
Initiates a new audio processing session. This event is emitted once when the session is created and ready.

### Method
SEND

### Endpoint
N/A (WebSocket Message)

### Payload
#### Properties
- **session_id** (string) - Required - UUID identifying the session.
- **created_at** (string) - Required - ISO 8601 timestamp in UTC.
- **type** (string) - Required - Lifecycle event emitted once when the session is created and ready. Must be 'start_session'.

### Request Example
```json
{
  "session_id": "550e8400-e29b-41d4-a716-446655440000",
  "created_at": "2025-09-19T12:34:00Z",
  "type": "start_session"
}
```
```

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

### On Speech Start Event

Source: https://docs.gladia.io/api-reference/v2/live/websocket

This event is sent when speech activity is detected on a channel. It includes session details and timing information.

```APIDOC
## On Speech Start Event

### Description
Indicates the start of detected speech activity on a given channel.

### Message Type
`send`

### Event Payload (JSON)
```json
{
  "session_id": "string",
  "created_at": "string",
  "type": "speech_start",
  "data": {
    "time": "number",
    "channel": "integer"
  }
}
```

### Parameters
#### Request Body
- **session_id** (string) - Required - UUID identifying the session.
- **created_at** (string) - Required - ISO 8601 timestamp in UTC.
- **type** (string) - Required - Must be `speech_start`.
- **data** (object) - Required - Timing information for speech start/end events.
  - **time** (number) - Optional - Time in seconds from session start when the event occurred.
  - **channel** (integer) - Optional - Audio channel index (0-based) for which the event applies.

### Example
```json
{
  "session_id": "550e8400-e29b-41d4-a716-446655440000",
  "created_at": "2025-09-19T12:34:06Z",
  "type": "speech_start",
  "data": {
    "time": 1.24,
    "channel": 0
  }
}
```
```

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

### POST /v2/live

Source: https://docs.gladia.io/api-reference/v2/live/init

Initiates a new live audio processing job. This endpoint allows you to start a real-time transcription session by providing details about the audio stream and processing preferences.

```APIDOC
## POST /v2/live

### Description
Initiates a new live job for audio processing.

### Method
POST

### Endpoint
https://api.gladia.io/v2/live

### Parameters
#### Query Parameters
- **region** (string) - Optional - The region used to process the audio. Supported values: `us-west`, `eu-west`.

#### Request Body
- **encoding** (string) - Optional - The encoding format of the audio stream. Supported formats: PCM (8, 16, 24, 32 bits), A-law (8 bits), μ-law (8 bits). Default: `wav/pcm`.
- **bit_depth** (integer) - Optional - The bit depth of the audio stream. Default: `16`.
- **sample_rate** (integer) - Optional - The sample rate of the audio stream. Default: `16000`.
- **channels** (integer) - Optional - The number of channels of the audio stream. Minimum: `1`, Maximum: `8`. Default: `1`.
- **custom_metadata** (object) - Optional - Custom metadata you can attach to this live transcription.
- **model** (string) - Optional - The model used to process the audio. Default: `solaria-1`.
- **endpointing** (number) - Optional - The endpointing duration in seconds. Default: `0.05`.
- **maximum_duration_without_endpointing** (number) - Optional - The maximum duration in seconds without endpointing. Default: `5`.
- **language_config** (object) - Optional - Specify the language configuration.
- **pre_processing** (object) - Optional - Specify the pre-processing configuration.
- **realtime_processing** (object) - Optional - Specify the realtime processing configuration.
- **post_processing** (object) - Optional - Specify the post-processing configuration.
- **messages_config** (object) - Optional - Specify the websocket messages configuration.
- **callback** (boolean) - Optional - If true, messages will be sent to configured url. Default: `false`.

### Request Example
```json
{
  "encoding": "wav/pcm",
  "sample_rate": 16000,
  "channels": 1,
  "custom_metadata": {
    "user": "John Doe"
  },
  "model": "solaria-1",
  "endpointing": 0.05,
  "maximum_duration_without_endpointing": 5,
  "language_config": {},
  "pre_processing": {},
  "realtime_processing": {},
  "post_processing": {},
  "messages_config": {},
  "callback": false
}
```

### Response
#### Success Response (201)
- **id** (string) - The unique identifier for the live job.
- **status** (string) - The status of the live job.

#### Response Example
```json
{
  "id": "a1b2c3d4-e5f6-7890-1234-567890abcdef",
  "status": "initiated"
}
```

#### Error Responses
- **400** - Bad Request: Something is wrong with the request.
- **401** - Unauthorized: You don't have the permissions to initiate a new live job.
- **422** - Unprocessable Entity: The parameters you gave are incorrect.
```

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

### Get Pre-recorded Jobs

Source: https://docs.gladia.io/api-reference/v2/pre-recorded/list

Fetches a list of pre-recorded jobs based on specified query parameters. Supports filtering by date, status, and custom metadata, as well as pagination.

```APIDOC
## GET /v2/pre-recorded

### Description
Get pre recorded jobs based on query parameters.

### Method
GET

### Endpoint
/v2/pre-recorded

### Parameters
#### Query Parameters
- **offset** (integer) - Optional - The starting point for pagination. A value of 0 starts from the first item. Defaults to 0.
- **limit** (integer) - Optional - The maximum number of items to return. Useful for pagination and controlling data payload size. Defaults to 20.
- **date** (string) - Optional - Filter items relevant to a specific date in ISO format (YYYY-MM-DD).
- **before_date** (string) - Optional - Include items that occurred before the specified date in ISO format.
- **after_date** (string) - Optional - Filter for items after the specified date. Use with `before_date` for a range. Date in ISO format.
- **status** (array of strings) - Optional - Filter the list based on item status. Accepts multiple values from the predefined list (queued, processing, done, error).
- **custom_metadata** (object) - Optional - Filter by custom metadata.

### Responses
#### Success Response (200)
- **first** (string) - URL to fetch the first page.
- **current** (string) - URL to fetch the current page.
- **next** (string) - URL to fetch the next page.
- **items** (array) - List of pre-recorded transcriptions.

#### Error Response (401)
- **timestamp** (string) - Date of when the error occurred.
- **path** (string) - Path to the API endpoint.
- **request_id** (string) - Debug id.
- **statusCode** (number) - HTTP status code of the error.
- **message** (string) - Error message.

### Security
- x_gladia_key
```

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

### Transcription Result Example

Source: https://docs.gladia.io/api-reference/v2/live/websocket

This example demonstrates the structure of a transcription result message. It contains word-level details including text, start and end times, and confidence scores, along with the overall transcribed text and detected language.

```json
{
  "session_id": "550e8400-e29b-41d4-a716-446655440000",
  "created_at": "2025-09-19T12:34:06Z",
  "type": "transcript",
  "data": {
    "words": [
      {
        "word": "Hello",
        "start": 0.24,
        "end": 0.36,
        "confidence": 0.91
      },
      {
        "word": "world",
        "start": 0.36,
        "end": 0.48,
        "confidence": 0.91
      }
    ],
    "text": "Hello world.",
    "language": "en"
  }
}
```

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

### Initiate Pre-recorded Transcription

Source: https://docs.gladia.io/api-reference/v2/transcription/init

Initiates a pre-recorded transcription job. Use the returned ID with the GET /v2/transcription/:id endpoint to get the results.

```APIDOC
## POST /v2/transcription/init

### Description
Initiates a pre-recorded transcription job. Use the returned ID with the GET /v2/transcription/:id endpoint to obtain the results.

### Method
POST

### Endpoint
/v2/transcription/init

### Parameters
#### Request Body
- **audio_url** (string) - Required - The URL of the audio file to transcribe.
- **webhook_url** (string) - Optional - The URL to send the transcription results to.
- **language** (string) - Optional - The language of the audio file (e.g., 'en', 'es', 'fr').

### Request Example
```json
{
  "audio_url": "https://example.com/audio.mp3",
  "webhook_url": "https://example.com/webhook",
  "language": "en"
}
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the transcription job.

#### Response Example
```json
{
  "id": "a1b2c3d4-e5f6-7890-1234-567890abcdef"
}
```
```

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

### Get Transcription by ID

Source: https://docs.gladia.io/api-reference/v2/transcription/get

Fetches the transcription details for a specific transcription ID.

```APIDOC
## GET /v2/transcription/{transcription_id}

### Description
Retrieves the transcription details for a given transcription ID.

### Method
GET

### Endpoint
/v2/transcription/{transcription_id}

### Parameters
#### Path Parameters
- **transcription_id** (string) - Required - The unique identifier of the transcription job.

### Response
#### Success Response (200)
- **id** (string) - The unique identifier of the transcription.
- **filename** (string) - The name of the uploaded file.
- **source** (string) - The link to download the file if an audio URL was used.
- **audio_duration** (number) - The duration of the audio file in seconds.
- **number_of_channels** (integer) - The number of channels in the audio file.

#### Response Example
```json
{
  "id": "45463597-20b7-4af7-b3b3-f5fb778203ab",
  "filename": "audio.mp3",
  "source": "http://example.com/audio.mp3",
  "audio_duration": 120.5,
  "number_of_channels": 2
}
```
```

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

### Audio to LLM Prompt Configuration

Source: https://docs.gladia.io/api-reference/v2/pre-recorded/list

Configure prompts and models for processing audio transcriptions with Large Language Models.

```APIDOC
## Audio to LLM Prompt Configuration

### Description
Configure prompts and models for processing audio transcriptions with Large Language Models.

### Parameters
#### Request Body
- **prompts** (array) - Required - The list of prompts applied on the audio transcription. Example: `["Extract the key points from the transcription"]`.
- **model** (string) - Optional - The model to use for the prompt execution. You can find the list of supported models [here](https://openrouter.ai/models). Defaults to 'openai/gpt-3.5-turbo'.
```

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

### Authenticate API Calls with API Key

Source: https://docs.gladia.io/api-reference/authentication

Use your API key in the 'x-gladia-key' header to authenticate all API requests. This method is universally applicable across all environments.

```bash
curl --request GET \
  --url https://api.gladia.io/v2/transcription \
  --header 'x-gladia-key: YOUR_GLADIA_API_KEY'
```

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

### Get Transcription Status and Results

Source: https://docs.gladia.io/api-reference/v2/transcription/get

Retrieve the status and results of a transcription job by its ID.

```APIDOC
## GET /v2/transcription/{transcriptionId}

### Description
Retrieves the status and results of a transcription job.

### Method
GET

### Endpoint
/v2/transcription/{transcriptionId}

### Parameters
#### Path Parameters
- **transcriptionId** (string) - Required - The ID of the transcription job to retrieve.

### Response
#### Success Response (200)
- **id** (string) - Id of the job
- **request_id** (string) - Debug id
- **version** (integer) - API version
- **status** (string) - Status of the job (queued, processing, done, error)
- **created_at** (string) - Creation date
- **completed_at** (string) - Completion date when status is "done" or "error"
- **custom_metadata** (object) - Custom metadata given in the initial request
- **error_code** (integer) - HTTP status code of the error if status is "error"
- **post_session_metadata** (object) - For debugging purposes, send data that could help to identify issues
- **kind** (string) - Type of transcription (pre-recorded or live)
- **file** (object) - The file data uploaded. Can be null if status is "error"
- **request_params** (object) - Parameters used for the transcription
- **result** (object) - Transcription's result when status is "done"

#### Response Example (Pre-recorded)
{
  "id": "45463597-20b7-4af7-b3b3-f5fb778203ab",
  "request_id": "G-45463597",
  "version": 2,
  "status": "done",
  "created_at": "2023-12-28T09:04:17.210Z",
  "completed_at": "2023-12-28T09:04:37.210Z",
  "custom_metadata": {
    "user": "John Doe"
  },
  "error_code": null,
  "post_session_metadata": {},
  "kind": "pre-recorded",
  "file": {
    "content_type": "audio/mpeg",
    "size": 1024000,
    "last_modified": "2023-12-28T09:04:17.210Z"
  },
  "request_params": {
    "language": "en"
  },
  "result": {
    "transcription": [
      {
        "alternatives": [
          {
            "content": "Hello world.",
            "confidence": 0.98
          }
        ],
        "start_time": 0.5,
        "end_time": 1.5
      }
    ],
    "language": "en",
    "language_is_accurate": true
  }
}

#### Response Example (Live)
{
  "id": "45463597-20b7-4af7-b3b3-f5fb778203ab",
  "request_id": "G-45463597",
  "version": 2,
  "status": "done",
  "created_at": "2023-12-28T09:04:17.210Z",
  "completed_at": "2023-12-28T09:04:37.210Z",
  "custom_metadata": {
    "user": "John Doe"
  },
  "error_code": null,
  "post_session_metadata": {},
  "kind": "live",
  "file": null,
  "request_params": {
    "language": "en"
  },
  "result": {
    "messages": [
      {
        "content": "Hello world.",
        "start_time": 0.5,
        "end_time": 1.5,
        "speaker": 0
      }
    ],
    "language": "en",
    "language_is_accurate": true
  }
}

#### Error Response (401)
{
  "timestamp": "2023-12-28T09:04:17.210Z",
  "path": "/v2/transcription/45463597-20b7-4af7-b3b3-f5fb778203ab",
  "request_id": "G-821fe9df",
  "statusCode": 401,
  "message": "gladia key not found"
}
```

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

### Download Live Audio File (OpenAPI)

Source: https://docs.gladia.io/api-reference/v2/live/get-audio

Use this OpenAPI definition to understand the request and response structure for downloading live audio. Ensure you include your API key in the request headers.

```yaml
GET /v2/live/{id}/file
openapi: 3.1.0
info:
  title: Gladia Control API
  description: ''
  version: '1.0'
  contact: {}
servers:
  - url: https://api.gladia.io/
    description: Gladia API production URL
security: []
tags: []
paths:
  /v2/live/{id}/file:
    get:
      tags:
        - Live V2
      summary: Download the audio file used for this live job
      operationId: StreamingController_getAudio_v2
      parameters:
        - name: id
          required: true
          in: path
          description: Id of the live job
          schema:
            example: 45463597-20b7-4af7-b3b3-f5fb778203ab
            type: string
      responses:
        '200':
          description: The audio file used for this live job
          content:
            application/octet-stream:
              schema:
                type: string
                format: binary
                example: <binary>
        '401':
          description: >-
            You don't have the permissions to access this live job or its audio
            file
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/UnauthorizedErrorResponse'
        '404':
          description: The live job or its audio file doesn't exist or has been deleted
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/NotFoundErrorResponse'
      security:
        - x_gladia_key: []
components:
  schemas:
    UnauthorizedErrorResponse:
      type: object
      properties:
        timestamp:
          type: string
          description: Date of when the error occurred
          example: '2023-12-28T09:04:17.210Z'
        path:
          type: string
          description: Path to the API endpoint
          example: /v2/transcription/45463597-20b7-4af7-b3b3-f5fb778203ab
        request_id:
          type: string
          description: Debug id
          example: G-821fe9df
        statusCode:
          type: number
          description: HTTP status code of the error
          example: 401
        message:
          type: string
          description: Error message
          example: gladia key not found
      required:
        - timestamp
        - path
        - request_id
        - statusCode
        - message
    NotFoundErrorResponse:
      type: object
      properties:
        timestamp:
          type: string
          description: Date of when the error occurred
          example: '2023-12-28T09:04:17.210Z'
        path:
          type: string
          description: Path to the API endpoint
          example: /v2/transcription/45463597-20b7-4af7-b3b3-f5fb778203ab
        request_id:
          type: string
          description: Debug id
          example: G-821fe9df
        statusCode:
          type: number
          description: HTTP status code of the error
          example: 404
        message:
          type: string
          description: Error message
          example: Not found
      required:
        - timestamp
        - path
        - request_id
        - statusCode
        - message
  securitySchemes:
    x_gladia_key:
      type: apiKey
      in: header
      name: x-gladia-key
      description: Your personal Gladia API key

```

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

### Callback Configuration

Source: https://docs.gladia.io/api-reference/v2/pre-recorded/init

Set up a callback URL to receive transcription results for pre-recorded audio.

```APIDOC
## CallbackConfigDto

### Description
Configuration for callback notifications.

### Properties
- **url** (string) - Required - The URL to be called with the result of the transcription.
  - Example: http://callback.example
- **method** (string) - Optional - The HTTP method to be used. Allowed values are `POST` or `PUT`. Defaults to `POST`.
  - Example: POST
```

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

### Get Transcription Job by ID

Source: https://docs.gladia.io/api-reference/v2/transcription/get

Fetches the metadata for a transcription job using its unique identifier.

```APIDOC
## GET /v2/transcription/{id}

### Description
Retrieves the metadata for a specific transcription job.

### Method
GET

### Endpoint
/v2/transcription/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - Id of the transcription job

### Responses
#### Success Response (200)
- **id** (string) - Id of the job
- **request_id** (string) - Debug id
- **version** (integer) - API version
- **kind** (string) - Type of transcription job (e.g., pre-recorded, live)
- **created_at** (string) - Creation date
- **status** (string) - Status of the job (queued, processing, done, error)
- **file** (object) - Details about the audio file
  - **id** (string) - File ID
  - **filename** (string) - Original filename
  - **source** (string) - URL of the audio source
  - **audio_duration** (integer) - Duration of the audio in seconds
  - **number_of_channels** (integer) - Number of audio channels
- **request_params** (object) - Parameters used for the transcription request
  - **audio_url** (string) - URL of the audio file
  - **subtitles** (boolean) - Whether subtitles were requested
  - **diarization** (boolean) - Whether diarization was enabled
  - **translation** (boolean) - Whether translation was requested
  - **summarization** (boolean) - Whether summarization was requested
  - **sentences** (boolean) - Whether sentence segmentation was requested
  - **moderation** (boolean) - Whether content moderation was enabled
  - **named_entity_recognition** (boolean) - Whether Named Entity Recognition was enabled
  - **name_consistency** (boolean) - Whether name consistency was enforced
  - **custom_spelling** (boolean) - Whether custom spelling was used
  - **structured_data_extraction** (boolean) - Whether structured data extraction was enabled
  - **chapterization** (boolean) - Whether chapterization was enabled
  - **sentiment_analysis** (boolean) - Whether sentiment analysis was enabled
  - **display_mode** (boolean) - Display mode setting
  - **audio_enhancer** (boolean) - Whether audio enhancement was used
  - **language_config** (object) - Language configuration
    - **code_switching** (boolean) - Whether code switching is enabled
    - **languages** (array) - List of languages detected or specified
  - **accurate_words_timestamps** (boolean) - Whether accurate word timestamps were generated
  - **diarization_enhanced** (boolean) - Whether enhanced diarization was used
  - **punctuation_enhanced** (boolean) - Whether enhanced punctuation was used
- **completed_at** (string) - Completion date (null if not completed)
- **custom_metadata** (object) - Custom metadata provided in the request (null if none)
- **error_code** (string) - Error code if the job failed (null if successful)
- **result** (object) - The transcription result (null if not available)

#### Error Response (401)
- Description: You don't have the permissions to access the transcription job.

#### Error Response (404)
- Description: The transcription job doesn't exist or has been deleted.

### Request Example
```json
{
  "example": ""
}
```

### Response Example
```json
{
  "id": "45463597-20b7-4af7-b3b3-f5fb778203ab",
  "request_id": "G-45463597",
  "version": 2,
  "kind": "pre-recorded",
  "created_at": "2023-12-28T09:04:17.210Z",
  "status": "queued",
  "file": {
    "id": "f0dcZE10-23d8-47f0-a25d-74a6eed88721",
    "filename": "split_infinity.wav",
    "source": "http://files.gladia.io/example/audio-transcription/split_infinity.wav",
    "audio_duration": 20,
    "number_of_channels": 1
  },
  "request_params": {
    "audio_url": "http://files.gladia.io/example/audio-transcription/split_infinity.wav",
    "subtitles": false,
    "diarization": false,
    "translation": false,
    "summarization": false,
    "sentences": false,
    "moderation": false,
    "named_entity_recognition": false,
    "name_consistency": false,
    "custom_spelling": false,
    "structured_data_extraction": false,
    "chapterization": false,
    "sentiment_analysis": false,
    "display_mode": false,
    "audio_enhancer": false,
    "language_config": {
      "code_switching": false,
      "languages": [
        "fr",
        "en"
      ]
    },
    "accurate_words_timestamps": false,
    "diarization_enhanced": false,
    "punctuation_enhanced": false
  },
  "completed_at": null,
  "custom_metadata": null,
  "error_code": null,
  "result": null
}
```
```

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

### Get Pre-recorded Job by ID

Source: https://docs.gladia.io/api-reference/v2/pre-recorded/get

Fetches the metadata for a pre-recorded job using its unique identifier.

```APIDOC
## GET /v2/pre-recorded/{id}

### Description
Retrieves the metadata for a pre-recorded job.

### Method
GET

### Endpoint
/v2/pre-recorded/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - Id of the pre recorded job

### Responses
#### Success Response (200)
- **PreRecordedResponse** - The pre recorded job's metadata

#### Error Responses
- **401** - UnauthorizedErrorResponse: You don't have the permissions to access the pre recorded job.
- **404** - NotFoundErrorResponse: The pre recorded job doesn't exist or has been deleted.
```

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

### Streaming Request Parameters

Source: https://docs.gladia.io/api-reference/v2/live/get

Configuration options for setting up a live audio stream for transcription.

```APIDOC
## Streaming Request Parameters

### Description
Configuration options for setting up a live audio stream for transcription.

### Properties

- **encoding** (string) - Optional - The encoding format of the audio stream. Supported formats: PCM (8, 16, 24, and 32 bits), A-law (8 bits), μ-law (8 bits). Default: `wav/pcm`.
- **bit_depth** (integer) - Optional - The bit depth of the audio stream. Default: `16`.
- **sample_rate** (integer) - Optional - The sample rate of the audio stream. Default: `16000`.
- **channels** (integer) - Optional - The number of channels of the audio stream. Minimum: `1`, Maximum: `8`. Default: `1`.
- **model** (string) - Optional - The model used to process the audio. Default: `solaria-1`.
- **endpointing** (number) - Optional - The endpointing duration in seconds. Minimum: `0.01`, Maximum: `10`. Default: `0.05`.
- **maximum_duration_without_endpointing** (number) - Optional - The maximum duration in seconds without endpointing. Minimum: `5`, Maximum: `60`. Default: `5`.
- **language_config** (object) - Optional - Specify the language configuration.
- **pre_processing** (object) - Optional - Specify the pre-processing configuration.
- **realtime_processing** (object) - Optional - Specify the realtime processing configuration.
- **post_processing** (object) - Optional - Specify the post-processing configuration.
- **messages_config** (object) - Optional - Specify the websocket messages configuration.
- **callback** (boolean) - Optional - If true, messages will be sent to configured url. Default: `false`.
- **callback_config** (object) - Optional - Specify the callback configuration.
```

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

### Get Transcription Details

Source: https://docs.gladia.io/api-reference/v2/transcription/get

Fetches the details of a specific transcription job. This includes configuration, status, and results.

```APIDOC
## GET /v2/transcription/{transcriptionId}

### Description
Retrieves the details of a transcription job, including its status, configuration, and results.

### Method
GET

### Endpoint
/v2/transcription/{transcriptionId}

### Parameters
#### Path Parameters
- **transcriptionId** (string) - Required - The unique identifier of the transcription job.

#### Query Parameters
- **language** (string) - Optional - The language of the transcription to retrieve. If not specified, the original transcription language will be returned.
- **with_words** (boolean) - Optional - If true, word-level details will be included in the response.
- **with_speakers** (boolean) - Optional - If true, speaker diarization information will be included in the response.
- **with_full_text** (boolean) - Optional - If true, the full text of the transcription will be included in the response.
- **with_audio_url** (boolean) - Optional - If true, a URL to the processed audio file will be included in the response.
- **with_subtitles** (string) - Optional - If provided with a valid format (srt, vtt), subtitles will be generated and included in the response.
- **with_translation** (string) - Optional - If provided with a valid language code, the transcription will be translated to the specified language.

### Response
#### Success Response (200)
- **id** (string) - The unique identifier of the transcription.
- **status** (string) - The current status of the transcription job (e.g., `processing`, `completed`, `failed`).
- **language** (string) - The language of the transcription.
- **words** (array) - An array of word objects, each containing the word, start time, and end time (if `with_words` is true).
- **speakers** (array) - An array of speaker objects, each containing speaker information (if `with_speakers` is true).
- **full_text** (string) - The full text of the transcription (if `with_full_text` is true).
- **audio_url** (string) - A URL to the processed audio file (if `with_audio_url` is true).
- **subtitles** (string) - The transcription formatted as subtitles (if `with_subtitles` is provided).
- **translation** (string) - The translated transcription (if `with_translation` is provided).

#### Response Example
```json
{
  "id": "a1b2c3d4-e5f6-7890-1234-567890abcdef",
  "status": "completed",
  "language": "en",
  "words": [
    {
      "word": "Hello",
      "start": 0.5,
      "end": 1.0
    },
    {
      "word": "world",
      "start": 1.1,
      "end": 1.5
    }
  ],
  "full_text": "Hello world",
  "audio_url": "https://api.gladia.io/v2/audio/processed/a1b2c3d4-e5f6-7890-1234-567890abcdef.mp3",
  "subtitles": "1\n00:00:00,500 --> 00:00:01,000\nHello\n\n2\n00:00:01,100 --> 00:00:01,500\nworld"
}
```
```

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

### InitStreamingResponse

Source: https://docs.gladia.io/api-reference/v2/live/init

This object represents the response received upon successfully initializing a live streaming session. It contains the necessary information to establish a WebSocket connection.

```APIDOC
## InitStreamingResponse

### Description
This object represents the response received upon successfully initializing a live streaming session. It contains the necessary information to establish a WebSocket connection.

### Properties
- **id** (string) - The unique identifier for the transcription job.
- **created_at** (string) - The timestamp indicating when the job was created.
- **url** (string) - The WebSocket URL to connect to for sending audio data. This URL includes a temporary token for authentication.
```

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

### Initialize Transcription

Source: https://docs.gladia.io/api-reference/v2/transcription/init

Initiates a new transcription job by providing the audio source and configuration options. The response contains the transcription job ID and a URL to fetch the results.

```APIDOC
## POST /v2/transcription

### Description
Initiates a new transcription job for an audio file. You can specify various configuration options to customize the transcription process, such as enabling sentiment analysis, PII redaction, or custom metadata.

### Method
POST

### Endpoint
/v2/transcription

### Parameters
#### Request Body
- **audio_url** (string) - Required - URL to a Gladia file or to an external audio or video file.
- **sentiment_analysis** (boolean) - Optional - Enable sentiment analysis for this audio. Defaults to false.
- **audio_to_llm** (boolean) - Optional - Enable audio to llm processing for this audio. Defaults to false.
- **audio_to_llm_config** (object) - Optional - Audio to llm configuration, if `audio_to_llm` is enabled.
- **pii_redaction** (boolean) - Optional - Enable PII redaction for this audio. Defaults to false.
- **pii_redaction_config** (object) - Optional - PII redaction configuration, if `pii_redaction` is enabled.
- **custom_metadata** (object) - Optional - Custom metadata you can attach to this transcription. Example: `{"user": "John Doe"}`.
- **sentences** (boolean) - Optional - Enable sentences for this audio. Defaults to false.
- **punctuation_enhanced** (boolean) - Optional - Use enhanced punctuation for this audio. Defaults to false.
- **language_config** (object) - Optional - Specify the language configuration.

### Request Example
```json
{
  "audio_url": "http://files.gladia.io/example/audio-transcription/split_infinity.wav",
  "sentiment_analysis": true,
  "pii_redaction": true,
  "custom_metadata": {
    "user": "John Doe"
  },
  "language_config": {
    "language": "en"
  }
}
```

### Response
#### Success Response (200)
- **id** (string) - Id of the job.
- **result_url** (string) - Prebuilt URL with your transcription `id` to fetch the result.

#### Response Example
```json
{
  "id": "45463597-20b7-4af7-b3b3-f5fb778203ab",
  "result_url": "https://api.gladia.io/v2/transcription/45463597-20b7-4af7-b3b3-f5fb778203ab"
}
```

### Error Handling
- **400 Bad Request**: Returned if the request is malformed or invalid.
- **401 Unauthorized**: Returned if the API key is invalid or missing.
- **422 Unprocessable Entity**: Returned if there are issues with the provided parameters, such as an invalid language code.
```

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

### On post chapterization

Source: https://docs.gladia.io/api-reference/v2/live/websocket

This message contains the informations about the post chapterization of the audio chunk.

```APIDOC
## PostChapterizationMessage

### Description
This message contains the informations about the post chapterization of the audio chunk.

### Type
`post_chapterization`

### Payload
```json
{
  "session_id": "string",
  "created_at": "string",
  "type": "post_chapterization",
  "data": {
    "metadata": {
      "audio_duration": "number",
      "number_of_distinct_channels": "number",
      "billing_time": "number",
      "transcription_time": "number"
    },
    "results": {
      "chapters": [
        {
          "start_time": "number",
          "end_time": "number",
          "summary": "string"
        }
      ]
    }
  }
}
```

### Example
```json
{
  "session_id": "550e8400-e29b-41d4-a716-446655440000",
  "created_at": "2025-09-19T12:45:00Z",
  "type": "post_chapterization",
  "data": {
    "metadata": {
      "audio_duration": 123.45,
      "number_of_distinct_channels": 1,
      "billing_time": 120,
      "transcription_time": 3.2
    },
    "results": {
      "chapters": [
        {
          "start_time": 0.5,
          "end_time": 10.2,
          "summary": "Introduction to the topic."
        },
        {
          "start_time": 10.5,
          "end_time": 25.8,
          "summary": "Detailed discussion of the main points."
        }
      ]
    }
  }
}
```
```

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

### Get Pre-recorded Audio Analysis

Source: https://docs.gladia.io/api-reference/v2/pre-recorded/get

Retrieve the analysis results for a pre-recorded audio file. This includes transcription, translation, and AI-generated insights.

```APIDOC
## GET /v2/pre-recorded/get

### Description
Retrieves the analysis results for a pre-recorded audio file. This endpoint is used to fetch the processed data after an audio file has been submitted for analysis.

### Method
GET

### Endpoint
/v2/pre-recorded/get

### Query Parameters
- **audio_url** (string) - Required - The URL of the pre-recorded audio file to analyze.
- **webhook_url** (string) - Optional - A URL to send the results to once processing is complete.
- **language** (string) - Optional - The language of the audio file (e.g., 'en', 'es', 'fr'). If not provided, the API will attempt to auto-detect the language.
- **translate_to** (string) - Optional - The language to translate the transcription into (e.g., 'en', 'es', 'fr'). Requires the `language` parameter to be set or auto-detected.
- **with_words** (boolean) - Optional - If true, includes word-level timestamps in the response.
- **with_sentences** (boolean) - Optional - If true, includes sentence-level segmentation in the response.
- **with_utterances** (boolean) - Optional - If true, includes speaker utterance segmentation in the response.
- **with_speakers** (boolean) - Optional - If true, includes speaker diarization information in the response.
- **with_emotions** (boolean) - Optional - If true, includes detected emotion analysis in the response.
- **with_named_entities** (boolean) - Optional - If true, includes named entity recognition in the response.
- **llm_prompt** (string) - Optional - A prompt to send to the LLM for analysis of the audio content.

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the analysis job.
- **status** (string) - The current status of the analysis job (e.g., 'processing', 'completed', 'failed').
- **results** (object) - Contains the analysis results once the job is completed. The structure of this object depends on the parameters requested.
  - **full_transcript** (string) - The complete transcription of the audio.
  - **words** (array) - Array of word objects with timestamps (if `with_words` is true).
  - **sentences** (array) - Array of sentence objects (if `with_sentences` is true).
  - **utterances** (array) - Array of utterance objects (if `with_utterances` is true).
  - **speakers** (array) - Array of speaker objects (if `with_speakers` is true).
  - **emotions** (array) - Array of emotion objects (if `with_emotions` is true).
  - **named_entities** (array) - Array of named entity objects (if `with_named_entities` is true).
  - **llm_response** (string) - The response from the LLM if `llm_prompt` was provided.

#### Error Response (400, 401, 403, 404, 429, 500)
- **error** (object) - Contains error details.
  - **status_code** (integer) - The HTTP status code of the error.
  - **exception** (string) - The type of exception that occurred.
  - **message** (string) - A detailed message explaining the error.

### Request Example
```bash
curl -X GET \
  'https://api.gladia.io/v2/pre-recorded/get?audio_url=YOUR_AUDIO_URL&with_words=true&llm_prompt=Summarize the key points of this audio.' \
  -H 'x-gladia-key: YOUR_API_KEY'
```

### Response Example (Success)
```json
{
  "id": "job_abc123",
  "status": "completed",
  "results": {
    "full_transcript": "Hello, this is a test transcript.",
    "words": [
      {"word": "Hello", "start": 0.5, "end": 1.0, "confidence": 0.95},
      {"word": "this", "start": 1.1, "end": 1.3, "confidence": 0.98}
    ],
    "llm_response": "The key points are about testing."
  }
}
```

### Response Example (Processing)
```json
{
  "id": "job_abc123",
  "status": "processing",
  "results": null
}
```
```