### Text Transcription Response Example

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This example shows the plain text response format for audio transcriptions. It provides only the transcribed text.

```text
And so, my fellow Americans, ask not what your country can do for you, ask what you can do for your country.
```

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

### VTT Transcription Response Example

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This example shows the WebVTT (VTT) format for audio transcriptions, commonly used for web video subtitles.

```text
WEBVTT

00:00:00.240 --> 00:00:07.839
And so, my fellow Americans, ask not what your country can do for you

00:00:07.839 --> 00:00:10.640
ask what you can do for your country.
```

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

### SRT Transcription Response Example

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This example demonstrates the SubRip Text (SRT) format for audio transcriptions, suitable for subtitle files.

```text
1
00:00:00,240 --> 00:00:07,839
And so, my fellow Americans, ask not what your country can do for you

2
00:00:07,839 --> 00:00:10,640
ask what you can do for your country.
```

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

### JSON Transcription Response Example

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This is an example of the default JSON response format for audio transcriptions. It contains the transcribed text.

```json
{
  "text": "Transcription text."
}
```

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

### Test Whisper Server API with Curl

Source: https://github.com/pfrankov/whisper-server/blob/main/AGENTS.md

Execute the API test suite using this script. Ensure the application is running before execution. This example demonstrates a POST request for audio transcription.

```bash
./test_api.sh
```

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions -F file=@jfk.wav -F response_format=json
```

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

### Verbose JSON Transcription Response Example

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This example shows the verbose JSON response format, which includes detailed information such as segments, timestamps, and probabilities.

```json
{
  "task": "transcribe",
  "language": "en",
  "duration": 10.5,
  "text": "Full transcription text.",
  "segments": [
    {
      "id": 0,
      "seek": 0,
      "start": 0.0,
      "end": 5.0,
      "text": "First segment.",
      "tokens": [50364, 13, 11, 263, 6116],
      "temperature": 0.0,
      "avg_logprob": -0.45,
      "compression_ratio": 1.275,
      "no_speech_prob": 0.1
    }
  ]
}
```

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

### List Available Models

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Retrieve a list of all available transcription models from both Whisper and Fluid providers. This GET request helps identify model IDs and defaults.

```bash
GET /v1/models
```

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

### Diarization JSON Response Example

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Example of a JSON response when diarization is enabled and 'response_format' is set to 'json'. It includes a 'speaker_segments' array detailing who spoke when.

```json
{
  "text": "Good morning everyone...",
  "speaker_segments": [
    {
      "speaker": "Speaker_1",
      "start": 0.0,
      "end": 4.2,
      "text": "Good morning everyone"
    },
    {
      "speaker": "Speaker_2",
      "start": 4.2,
      "end": 7.8,
      "text": "Morning! Shall we begin?"
    }
  ]
}
```

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

### Build Whisper Server Application

Source: https://github.com/pfrankov/whisper-server/blob/main/AGENTS.md

Use this command to build the Whisper Server application from the command line.

```bash
xcodebuild build -project WhisperServer.xcodeproj -scheme WhisperServer
```

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

### Cloning WhisperServer Repository

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Commands to clone the WhisperServer repository from GitHub and navigate into the project directory. This is the first step to building from source.

```bash
git clone https://github.com/pfrankov/whisper-server.git
cd whisper-server
```

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

### Generate App Icon

Source: https://github.com/pfrankov/whisper-server/blob/main/AGENTS.md

Run this script to generate the application icon using a source image.

```bash
./GenerateAppIcon.sh
```

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

### Run All API Tests

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Execute all automated API tests for WhisperServer. Ensure the server is running and a test audio file is present.

```bash
./test_api.sh
```

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

### Basic Transcription (JSON)

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Perform a basic audio transcription using the default JSON response format. Ensure the audio file is in the project root.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@jfk.wav
```

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

### List Available Models

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Retrieves a list of available audio models supported by WhisperServer. This is useful for checking model IDs and default configurations.

```APIDOC
## GET /v1/models

### Description
Lists available models from both Whisper and Fluid providers. Useful to check IDs and defaults.

### Method
GET

### Endpoint
/v1/models
```

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

### Transcribe Audio via HTTP POST

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Use this cURL command to send an audio file for transcription to the local WhisperServer. Ensure the file path is correct.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@/path/to/audio.mp3
```

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

### API Key Authentication Request

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This command demonstrates how to authenticate API requests using a bearer token when 'Require API Key' is enabled. Replace '<your-key>' with your actual generated API key.

```bash
curl -X POST http://192.168.1.42:12017/v1/audio/transcriptions \
  -H "Authorization: Bearer ws-<your-key>" \
  -F file=@audio.wav
```

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

### List Available Test Groups

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

List the available test groups for targeted execution. Use this to understand the scope of different test categories.

```bash
./test_api.sh --list-groups
```

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

### Run Specific Test Groups: Models and Negative

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Execute tests for both the 'models' and 'negative' test groups. This allows for combined testing of model listing and error handling.

```bash
./test_api.sh --only=models,negative
```

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

### Streaming Transcription (SSE, JSON)

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Enable streaming transcription with Server-Sent Events (SSE) and receive JSON chunks. Requires 'Accept: text/event-stream' header and 'stream=true'.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -H "Accept: text/event-stream" \
  -F file=@jfk.wav \
  -F response_format=json \
  -F stream=true \
  --no-buffer
```

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

### Streaming Transcription with SSE

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Use this command to stream audio transcriptions using Server-Sent Events (SSE). Ensure the client sends the 'Accept: text/event-stream' header.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -H "Accept: text/event-stream" \
  -F file=@audio.wav \
  -F stream=true \
  --no-buffer
```

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

### Streaming Transcription (Chunked, Text)

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Enable streaming transcription with HTTP chunked encoding and receive text chunks. Use 'stream=true' and 'response_format=text'.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@jfk.wav \
  -F response_format=text \
  -F stream=true \
  --no-buffer
```

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

### Streaming Transcription with Chunked Transfer

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This command is used for streaming audio transcriptions when SSE is not supported, falling back to HTTP chunked transfer encoding.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@audio.wav \
  -F stream=true \
  --no-buffer
```

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

### Transcription with VTT Format

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Perform an audio transcription and receive the output in VTT subtitle format. Use 'response_format=vtt' for this.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@jfk.wav \
  -F response_format=vtt
```

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

### Diarization with Fluid Provider

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Enable speaker diarization for transcriptions using the Fluid provider. This adds 'speaker_segments' to the JSON output. Requires a Fluid model and 'diarize=true'.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@jfk.wav \
  -F model=parakeet-tdt-0.6b-v3 \
  -F response_format=json \
  -F diarize=true
```

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

### Transcription with Text Format

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Perform an audio transcription and receive the output in plain text format. Specify 'response_format=text' to enable this.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@jfk.wav \
  -F response_format=text
```

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

### Audio Transcription with SSE Streaming

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Transcribe audio in real-time using Server-Sent Events (SSE) for streaming. This method is preferred when the client supports SSE.

```APIDOC
## POST /v1/audio/transcriptions (SSE Streaming)

### Description
Transcribes audio content in real-time using Server-Sent Events (SSE) for streaming.

### Method
POST

### Endpoint
/v1/audio/transcriptions

### Headers
- **Accept**: `text/event-stream`

### Parameters
#### Form Data
- **file** (file) - Required - The audio file to transcribe.
- **stream** (boolean) - Required - Set to `true` to enable streaming.

### Request Example
```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -H "Accept: text/event-stream" \
  -F file=@audio.wav \
  -F stream=true \
  --no-buffer
```

### Response
#### Success Response (200)
Streaming response with `data:` lines for transcribed segments and an `event: end` for completion.

#### Response Example
```
data: First transcribed segment
data:

data: Second transcribed segment
data:

event: end
data:
```
```

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

### Run Specific Test Group: Models

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Execute only the tests related to the /v1/models endpoint. Useful for isolating issues with model listing.

```bash
./test_api.sh --only=models
```

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

### Transcription with Verbose JSON

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Perform an audio transcription and receive detailed JSON output, including segment information. Use 'response_format=verbose_json'.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@jfk.wav \
  -F response_format=verbose_json
```

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

### API Key Authentication

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Secure API access using an API key. Clients must include the key in the `Authorization` header.

```APIDOC
## API Key Authentication

### Description
When 'Require API Key' is enabled, clients must authenticate requests by providing a bearer token in the `Authorization` header.

### Headers
- **Authorization**: `Bearer ws-<your-api-key>`

### Request Example
```bash
curl -X POST http://192.168.1.42:12017/v1/audio/transcriptions \
  -H "Authorization: Bearer ws-<your-key>" \
  -F file=@audio.wav
```

### Key Management
- **Copy API Key**: Copies the current API key to the clipboard.
- **Regenerate API Key…**: Generates a new API key, invalidating the previous one. The new key is automatically copied to the clipboard.
```

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

### Audio Transcription with Chunked Transfer Encoding

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Transcribe audio with streaming fallback using HTTP chunked transfer encoding when SSE is not supported.

```APIDOC
## POST /v1/audio/transcriptions (Chunked Encoding)

### Description
Transcribes audio content with streaming fallback using HTTP chunked transfer encoding when Server-Sent Events (SSE) are not supported.

### Method
POST

### Endpoint
/v1/audio/transcriptions

### Parameters
#### Form Data
- **file** (file) - Required - The audio file to transcribe.
- **stream** (boolean) - Required - Set to `true` to enable streaming.

### Request Example
```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@audio.wav \
  -F stream=true \
  --no-buffer
```

### Response
#### Success Response (200)
HTTP chunked transfer encoding response containing transcribed segments.
```

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

### Transcription with SRT Format

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Perform an audio transcription and receive the output in SRT subtitle format. Set 'response_format=srt' for this output.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@jfk.wav \
  -F response_format=srt
```

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

### Audio Transcription with Diarization

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Transcribe audio and enable speaker diarization to identify different speakers. Requires the Fluid provider.

```APIDOC
## POST /v1/audio/transcriptions (Diarization)

### Description
Transcribes audio content and enables speaker diarization to identify who is speaking. This feature requires the Fluid provider.

### Method
POST

### Endpoint
/v1/audio/transcriptions

### Parameters
#### Form Data
- **file** (file) - Required - The audio file to transcribe.
- **model** (string) - Optional - The model ID to use (e.g., `parakeet-tdt-0.6b-v3`).
- **response_format** (string) - Optional - The desired response format. Supported values include `json` and `verbose_json`.
- **diarize** (boolean) - Required - Set to `true` to enable diarization.

### Request Example
```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@meeting.wav \
  -F model=parakeet-tdt-0.6b-v3 \
  -F response_format=json \
  -F diarize=true
```

### Response
#### Success Response (200)
- **text** (string) - The transcribed text.
- **speaker_segments** (array) - An array of objects, each representing a segment with speaker information.
  - **speaker** (string) - The identified speaker (e.g., `Speaker_1`).
  - **start** (number) - The start time of the segment in seconds.
  - **end** (number) - The end time of the segment in seconds.
  - **text** (string) - The transcribed text for this segment.

#### Response Example (JSON)
```json
{
  "text": "Good morning everyone...",
  "speaker_segments": [
    {
      "speaker": "Speaker_1",
      "start": 0.0,
      "end": 4.2,
      "text": "Good morning everyone"
    },
    {
      "speaker": "Speaker_2",
      "start": 4.2,
      "end": 7.8,
      "text": "Morning! Shall we begin?"
    }
  ]
}
```

### Streaming with Diarization
When streaming is enabled with diarization, a single JSON chunk containing `speaker_segments` is sent upon completion, followed by the standard `end` event.
```

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

### Audio Transcription Endpoint

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This endpoint accepts audio files for transcription and returns the transcribed text in various formats.

```APIDOC
## POST /v1/audio/transcriptions

### Description
Transcribes an audio file. The server is compatible with the OpenAI Whisper API.

### Method
POST

### Endpoint
http://localhost:12017/v1/audio/transcriptions

### Parameters
#### Path Parameters
None

#### Query Parameters
- **model** (string) - Optional - Model to use. Accepts model IDs.
- **prompt** (string) - Optional - Guides the style or tone of the transcription (Whisper specific).
- **response_format** (string) - Optional - Specifies the output format. Values: json, text, srt, vtt, verbose_json. Defaults to json.
- **language** (string) - Optional - The input language, specified as a 2-letter ISO 639-1 code.
- **diarize** (boolean) - Optional - Enables Fluid speaker diarization. Defaults to false.
- **stream** (boolean) - Optional - Enables streaming output via Server-Sent Events (SSE) or chunked transfer. Defaults to false.

#### Request Body
- **file** (file) - Required - The audio file to transcribe. Supported formats: wav, mp3, m4a.

### Request Example
```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@/path/to/audio.mp3 \
  -F response_format=json \
  -F language=en
```

### Response
#### Success Response (200)
- **text** (string) - The transcribed text (for json, text response formats).
- **task** (string) - The type of task performed (e.g., "transcribe") (for verbose_json format).
- **language** (string) - The detected language of the audio (for verbose_json format).
- **duration** (number) - The duration of the audio in seconds (for verbose_json format).
- **segments** (array) - An array of segment objects, each containing details like start time, end time, and text (for verbose_json format).

#### Response Example (json)
```json
{
  "text": "Transcription text."
}
```

#### Response Example (verbose_json)
```json
{
  "task": "transcribe",
  "language": "en",
  "duration": 10.5,
  "text": "Full transcription text.",
  "segments": [
    {
      "id": 0,
      "seek": 0,
      "start": 0.0,
      "end": 5.0,
      "text": "First segment.",
      "tokens": [50364, 13, 11, 263, 6116],
      "temperature": 0.0,
      "avg_logprob": -0.45,
      "compression_ratio": 1.275,
      "no_speech_prob": 0.1
    }
  ]
}
```

#### Response Example (text)
```
And so, my fellow Americans, ask not what your country can do for you, ask what you can do for your country.
```

#### Response Example (srt)
```
1
00:00:00,240 --> 00:00:07,839
And so, my fellow Americans, ask not what your country can do for you

2
00:00:07,839 --> 00:00:10,640
ask what you can do for your country.
```

#### Response Example (vtt)
```
WEBVTT

00:00:00.240 --> 00:00:07.839
And so, my fellow Americans, ask not what your country can do for you

00:00:07.839 --> 00:00:10.640
ask what you can do for your country.
```
```

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

### Transcribe Audio with JSON Response Format

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

This cURL command demonstrates how to request transcription results in JSON format. The server defaults to JSON if no format is specified.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@/path/to/audio.mp3 \
  -F response_format=json
```

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

### Run Specific Test Group: Fluid

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Execute only the tests related to the Fluid transcription provider. This is useful for testing diarization and Fluid-specific features.

```bash
./test_api.sh --only=fluid
```

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

### Curl Request for Audio Transcription

Source: https://github.com/pfrankov/whisper-server/blob/main/AGENTS.md

Manually test audio transcription endpoints using curl. Supports various response formats like JSON, text, SRT, and VTT.

```bash
curl -F file=@jfk.wav -F response_format=json http://localhost:12017/v1/audio/transcriptions
```

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

### Audio Transcriptions

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Transcribes an audio file using a specified model. Supports various response formats and streaming.

```APIDOC
## POST /v1/audio/transcriptions

### Description
Transcribes an audio file. Supports multiple response formats and streaming.

### Method
POST

### Endpoint
/v1/audio/transcriptions

### Parameters
#### Request Body
- **file** (file) - Required - The audio file to transcribe (e.g., wav, mp3, m4a).
- **model** (string) - Optional - The ID of the model to use for transcription.
- **response_format** (string) - Optional - The desired format for the transcription response. Options include `json` (default), `text`, `srt`, `vtt`, `verbose_json`.
- **language** (string) - Optional - The ISO-639-1 code for the language of the audio (e.g., `en`).
- **prompt** (string) - Optional - A text prompt to guide the transcription.
- **stream** (boolean) - Optional - Enables streaming of transcription results (`true`/`false`).
- **diarize** (boolean) - Optional - Enables speaker diarization for the Fluid provider only (`true`/`false`). If true, `speaker_segments` will be included in JSON responses.

### Response
Content-Type depends on `response_format`. SSE responses use `text/event-stream` and end with an `end` event.
```

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

### Diarization Enabled Transcription

Source: https://github.com/pfrankov/whisper-server/blob/main/README.md

Enable speaker diarization by setting 'diarize=true' in your request when using the FluidAudio provider. This command also specifies a Fluid model and JSON response format.

```bash
curl -X POST http://localhost:12017/v1/audio/transcriptions \
  -F file=@meeting.wav \
  -F model=parakeet-tdt-0.6b-v3 \
  -F response_format=json \
  -F diarize=true
```

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

### Run Specific Test Group: Whisper

Source: https://github.com/pfrankov/whisper-server/blob/main/TESTING.md

Execute only the tests related to the core Whisper transcription functionality. This group focuses on the primary transcription endpoint.

```bash
./test_api.sh --only=whisper
```

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