pyannoteAI

https://docs.pyannote.ai/
Admin
pyannoteAI is an AI platform for speaker diarization that identifies who spoke when in audio files,...
Tokens:78,725
Snippets:551
Trust Score:7.4
Update:1 day ago
Show doc for...
Tokens:
Raw
### Go GET request to retrieve job by ID

Source: https://docs.pyannote.ai/api-reference/get-job

Go example using net/http to GET a job by ID. Prints the response body.

```go
package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.pyannote.ai/v1/jobs/{jobId}"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("Authorization", "Bearer <token>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}
```

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

### Go: Get all jobs

Source: https://docs.pyannote.ai/api-reference/list-jobs

Uses net/http to send a GET request and print the response body.

```go
package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.pyannote.ai/v2/jobs"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("Authorization", "Bearer <token>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}
```

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

### Ruby GET request to retrieve job by ID

Source: https://docs.pyannote.ai/api-reference/get-job

Ruby example using net/http to GET a job by ID. Prints the response body.

```ruby
require 'uri'
require 'net/http'

url = URI("https://api.pyannote.ai/v1/jobs/{jobId}")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)
request["Authorization"] = 'Bearer <token>'

response = http.request(request)
puts response.read_body
```

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

### Python GET request to retrieve job by ID

Source: https://docs.pyannote.ai/api-reference/get-job

Python example using the requests library to GET a job by ID. Requires the requests package and a valid API token.

```python
import requests

url = "https://api.pyannote.ai/v1/jobs/{jobId}"

headers = {"Authorization": "Bearer <token>"}

response = requests.get(url, headers=headers)

print(response.text)
```

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

### Example OpenAI transcript segments

Source: https://docs.pyannote.ai/tutorials/diarization-asr-merge

Example JSON output from OpenAI gpt-4o-transcribe or whisper-1 API showing transcript segments with start and end timestamps in seconds.

```json
{
  "task": "transcribe",
  "duration": 42.7,
  "text": "Agent: Thanks for calling OpenAI support.\nCustomer: Hi, I need help with diarization.",
  "segments": [
    {
      "type": "transcript.text.segment",
      "id": "seg_001",
      "start": 0.0,
      "end": 5.2,
      "text": "Thanks for calling OpenAI support."
    },
    {
      "type": "transcript.text.segment",
      "id": "seg_002",
      "start": 5.2,
      "end": 12.8,
      "text": "Hi, I need help with diarization."
    }
  ],
  "usage": {
    "type": "duration",
    "seconds": 43
  }
}
```

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

### Java: Get all jobs

Source: https://docs.pyannote.ai/api-reference/list-jobs

Uses Unirest to send a GET request and get the response as a string.

```java
HttpResponse<String> response = Unirest.get("https://api.pyannote.ai/v2/jobs")
  .header("Authorization", "Bearer <token>")
  .asString();
```

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

### Example response from identify endpoint

Source: https://docs.pyannote.ai/tutorials/identification-with-voiceprints

Example JSON response showing a jobId and status 'created' for tracking the identification job.

```JSON
{
  "jobId": "4d9b9ab6-edd7-5feca-b86e-gee75840775e",
  "status": "created"
}
```

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

### GET /v1/test

Source: https://docs.pyannote.ai/authentication

Example request demonstrating authentication with a Bearer token. This endpoint is used to test authentication.

```APIDOC
## GET /v1/test

### Description
Example request demonstrating authentication with a Bearer token. This endpoint is used to test authentication.

### Method
GET

### Endpoint
https://api.pyannote.ai/v1/test

### Authentication
Bearer token required. Include the API key in the Authorization header.

### Request Example
```
curl --request GET \
  --url https://api.pyannote.ai/v1/test \
  --header: 'Authorization: Bearer <apikey>' \
  --header: 'Content-Type: application/json'
```

### Response
No response details provided in the source.
```

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

### Python: Get all jobs

Source: https://docs.pyannote.ai/api-reference/list-jobs

Uses the requests library to send a GET request and print the response.

```python
import requests

url = "https://api.pyannote.ai/v2/jobs"

headers = {"Authorization": "Bearer <token>"}

response = requests.get(url, headers=headers)

print(response.text)
```

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

### Ruby: Get all jobs

Source: https://docs.pyannote.ai/api-reference/list-jobs

Uses Net::HTTP to send a GET request and print the response body.

```ruby
require 'uri'
require 'net/http'

url = URI("https://api.pyannote.ai/v2/jobs")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)
request["Authorization"] = 'Bearer <token>'

response = http.request(request)
puts response.read_body
```

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

### Go GET request to retrieve live stream

Source: https://docs.pyannote.ai/api-reference/get-stream

Uses net/http to send a GET request with an Authorization header and prints the response body.

```go
package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.pyannote.ai/v1/live/{id}"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("Authorization", "Bearer <token>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}
```

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

### Go GET request to test endpoint

Source: https://docs.pyannote.ai/api-reference/test

Uses net/http to send a GET request with Bearer token authorization and prints the response body.

```go
package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.pyannote.ai/v1/test"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("Authorization", "Bearer <token>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}
```

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

### Example formatted transcript output

Source: https://docs.pyannote.ai/tutorials/speech-to-text-diarization

Example output produced by the formatting functions, showing two turns with speaker labels and timestamps.

```Plain Text
SPEAKER_00 (0:00): Hello everyone, welcome to the meeting.
SPEAKER_01 (0:03): Hi, thanks for having me.
```

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

### 200 OK response body

Source: https://docs.pyannote.ai/api-reference/get-stream

Example JSON response for a successful request.

```json
{
  "id": "<string>",
  "startedAt": "2023-11-07T05:31:56Z",
  "completedAt": "2023-11-07T05:31:56Z",
  "input": {},
  "status": "created"
}
```

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

### Response body: 201 success

Source: https://docs.pyannote.ai/api-reference/upload-media

Example of a successful response containing the upload URL.

```json
{
  "url": "<string>"
}

```

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

### Main async workflow

Source: https://docs.pyannote.ai/tutorials/live-diarized-transcription-per-speaker-stream

Sets up signal handling, creates the pyannote stream, builds the AssemblyAI WebSocket URL with query parameters, and opens all WebSocket connections using an AsyncExitStack.

```python
async def main() -> None:
    loop = asyncio.get_running_loop()
    stop = asyncio.Event()
    loop.add_signal_handler(signal.SIGINT, stop.set)

    stream_id, pyannote_url = create_pyannote_stream()
    print(f"pyannoteAI stream: {stream_id}")

    params = urlencode({
        "sample_rate": SAMPLE_RATE,
        "encoding": "pcm_s16le",
        "speech_model": "universal-3-5-pro",
        "speaker_labels": "false",
        "min_turn_silence": 100,
        "max_turn_silence": 1000,
        "continuous_partials": "true",
    })
    assembly_url = f"wss://streaming.assemblyai.com/v3/ws?{params}"

    async with AsyncExitStack() as stack:
        pyannote_stream = await stack.enter_async_context(websockets.connect(pyannote_url))
        assembly_streams = [
            await stack.enter_async_context(
                websockets.connect(
                    assembly_url,
                    additional_headers={"Authorization": ASSEMBLYAI_API_KEY},
                )
            )
            for _ in range(max_speakers)
        ]
```

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

### Response: 200 OK

Source: https://docs.pyannote.ai/api-reference/list-jobs

Example of a successful response containing a list of jobs and a nextCursor.

```json
{
  "items": [
    {
      "id": "<string>",
      "status": "<string>",
      "createdAt": "2023-11-07T05:31:56Z"
    }
  ],
  "nextCursor": "a1b2c3d4-e5f6-7890-abcd-ef1234567890"
}
```

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

### Java Unirest GET request to retrieve job by ID

Source: https://docs.pyannote.ai/api-reference/get-job

Java example using Unirest to GET a job by ID. Returns the response as a string.

```java
HttpResponse<String> response = Unirest.get("https://api.pyannote.ai/v1/jobs/{jobId}")
  .header("Authorization", "Bearer <token>")
  .asString();
```

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

### Upload audio file with Python (requests)

Source: https://docs.pyannote.ai/tutorials/how-to-upload-files

Uses requests to create a pre-signed PUT URL and upload a local file. The object-key is unique to your team. In production, use environment variables for the API key.

```Python
import requests

# Define your media object key
object_key = "my-meeting-recording"  # Replace with your desired object-key

# Create the pre-signed PUT URL.
api_key = "YOUR_API_KEY"  # In production, use environment variables: os.getenv("PYANNOTE_API_KEY")
response = requests.post(
    "https://api.pyannote.ai/v1/media/input",
    json={"url": f"media://{object_key}"},
    headers={
        "Authorization": f"Bearer {api_key}",
        "Content-Type": "application/json"
    }
)
response.raise_for_status()
data = response.json()
presigned_url = data["url"]

# Upload local file to the pre-signed URL.
print("Uploading {0} to {1}".format(input_path, presigned_url))
with open(input_path, "rb") as input_file:
    # Upload your local audio file.
    requests.put(presigned_url, data=input_file)
```

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

### PHP cURL GET request to retrieve job by ID

Source: https://docs.pyannote.ai/api-reference/get-job

PHP example using cURL to GET a job by ID. Prints the response or a cURL error message.

```php
<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.pyannote.ai/v1/jobs/{jobId}",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}
```

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

### Create .env file with API keys

Source: https://docs.pyannote.ai/tutorials/live-diarized-transcription-single-stt-stream

Environment file containing the required API keys: PYANNOTEAI_API_KEY and OPENAI_API_KEY. Used by the script to authenticate with both services.

```plaintext
PYANNOTEAI_API_KEY=sk_xxx
OPENAI_API_KEY=sk-xxx
```

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

### JavaScript fetch GET request to retrieve job by ID

Source: https://docs.pyannote.ai/api-reference/get-job

JavaScript example using fetch to GET a job by ID. Logs the response JSON or error to the console.

```javascript
const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.pyannote.ai/v1/jobs/{jobId}', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));
```

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

### Go example for diarize

Source: https://docs.pyannote.ai/api-reference/diarize

Uses net/http to POST the diarize payload as a JSON string. Requires a bearer token and prints the response body.

```go
package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.pyannote.ai/v1/diarize"

	payload := strings.NewReader("{\n  \"url\": \"https://example.com/audio.wav\",\n  \"webhook\": \"https://example.com/webhook\",\n  \"webhookStatusOnly\": true,\n  \"model\": \"precision-2\",\n  \"numSpeakers\": 2,\n  \"minSpeakers\": 1,\n  \"maxSpeakers\": 4,\n  \"vadSensitivity\": 0,\n  \"crosstalkSensitivity\": 0,\n  \"turnLevelConfidence\": true,\n  \"speechProbability\": true,\n  \"crosstalkProbability\": true,\n  \"speakerProbability\": true,\n  \"exclusive\": true,\n  \"confidence\": true,\n  \"transcription\": false,\n  \"transcriptionConfig\": {\n    \"model\": \"faster-whisper-large-v3-turbo\"\n  }\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}
```

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

### Go: Get upload URL

Source: https://docs.pyannote.ai/api-reference/upload-media

Uses net/http to POST the media URL as a JSON string. Prints the response body.

```go
package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.pyannote.ai/v1/media/input"

	payload := strings.NewReader("{\n  \"url\": \"<string>\"\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

```

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

### Merged diarization + ASR segments

Source: https://docs.pyannote.ai/tutorials/diarization-asr-merge

Example output showing the resulting merged segments with start, end, text, and assigned speaker labels.

```json
[
  {
    "start": 0.0,
    "end": 5.2,
    "text": "Thanks for calling OpenAI support.",
    "speaker": "SPEAKER_00"
  },
  {
    "start": 5.2,
    "end": 12.8,
    "text": "Hi, I need help with diarization.",
    "speaker": "SPEAKER_01"
  }
]
```

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

### Environment variables for API keys

Source: https://docs.pyannote.ai/tutorials/live-diarized-transcription-per-speaker-stream

Create a .env file with your pyannoteAI and AssemblyAI API keys.

```bash
PYANNOTEAI_API_KEY=sk_xxx
ASSEMBLYAI_API_KEY=xxxxxxxx
```

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

### 200 Success response schema for Get Job by ID

Source: https://docs.pyannote.ai/api-reference/get-job

Example of a successful 200 response containing job details, including status, timestamps, and output fields such as diarization, confidence, and transcriptions.

```json
{
  "jobId": "fb16c565-f3f0-4402-a08c-9d44df0ccc7b",
  "status": "pending",
  "createdAt": "2024-02-20T12:00:00Z",
  "updatedAt": "2024-02-20T12:00:00Z",
  "output": {
    "diarization": [
      {
        "speaker": "SPEAKER_00",
        "start": 15,
        "end": 30.5,
        "confidence": {
          "SPEAKER_00": 16,
          "SPEAKER_01": 93
        }
      }
    ],
    "confidence": {
      "score": [
        95,
        89,
        78,
        67,
        56,
        45,
        34,
        23,
        12,
        1
      ],
      "resolution": 0.02
    },
    "exclusiveDiarization": [
      {
        "speaker": "SPEAKER_00",
        "start": 15,
        "end": 30.5,
        "confidence": {
          "SPEAKER_00": 16,
          "SPEAKER_01": 93
        }
      }
    ],
    "speechProbability": {
      "score": [
        95,
        89,
        78,
        67,
        56,
        45,
        34,
        23,
        12,
        1
      ],
      "resolution": 0.02
    },
    "crosstalkProbability": {
      "score": [
        3,
        8,
        61,
        74,
        12,
        4,
        2,
        1,
        0,
        0
      ],
      "resolution": 0.02
    },
    "speakerProbability": {
      "score": {
        "SPEAKER_00": [
          95,
          89,
          12
        ],
        "SPEAKER_01": [
          5,
          11,
          88
        ]
      },
      "resolution": 0.02
    },
    "wordLevelTranscription": [
      {
        "start": 0.5,
        "end": 0.8,
        "text": "Hello",
        "speaker": "SPEAKER_00"
      }
    ],
    "turnLevelTranscription": [
      {
        "start": 0.5,
        "end": 2.3,
        "text": "Hello, how are you?",
        "speaker": "SPEAKER_00"
      }
    ],
    "error": "<string>",
    "warning": "<string>"
  }
}
```

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

### Main async pipeline for live diarized transcription

Source: https://docs.pyannote.ai/tutorials/live-diarized-transcription-per-speaker-stream

Sets up an asyncio-based pipeline that captures microphone audio, sends it to Pyannote and multiple AssemblyAI streams, and coordinates results using a SpeakerRouter and TranscriptTimeline. The main function waits for a stop event, then gracefully shuts down by stopping the mic, flushing the router, and cancelling tasks.

```python
queue: "asyncio.Queue[np.ndarray]" = asyncio.Queue(maxsize=50)
        router = SpeakerRouter()
        timeline = TranscriptTimeline()
        mic = microphone(loop, queue)
        timeline.render()

        pyannote_done = asyncio.Event()
        assembly_done = [asyncio.Event() for _ in assembly_streams]
        pyannote_task = asyncio.create_task(
            receive_pyannote(pyannote_stream, router, stop, pyannote_done)
        )
        assembly_tasks = [
            asyncio.create_task(
                receive_assemblyai(
                    stream,
                    index,
                    router,
                    timeline,
                    stop,
                    assembly_done[index],
                )
            )
            for index, stream in enumerate(assembly_streams)
        ]
        pump_task = asyncio.create_task(
            pump_audio(queue, pyannote_stream, assembly_streams, router, stop)
        )
        tasks = [pyannote_task, *assembly_tasks, pump_task]

        await stop.wait()
        mic.stop()
        mic.close()
        await pump_task
        await pyannote_stream.send(json.dumps({"type": "end_of_stream"}))
        await pyannote_done.wait()
        await router.flush(assembly_streams, force=True)
        for stream in assembly_streams:
            await stream.send(json.dumps({"type": "Terminate"}))
        await asyncio.gather(*(done.wait() for done in assembly_done))
        for task in tasks:
            task.cancel()
        await asyncio.gather(*tasks, return_exceptions=True)


if __name__ == "__main__":
    asyncio.run(main())
```

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

### Webhook payload for a succeeded diarization job

Source: https://docs.pyannote.ai/webhooks/receiving-webhooks

Example request body sent when a diarization job succeeds. The output field contains a diarization schema with segments, each having start, end, and speaker fields.

```json
{
  "jobId": "job_id",
  "status": "succeeded",
  "output": {
    "diarization": [
      {
        "start": 0.0,
        "end": 1.0,
        "speaker": "speaker_1"
      }
      ...
    ]
  }
}
```

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

### Go: Create live stream

Source: https://docs.pyannote.ai/api-reference/create-stream

Uses net/http to POST an empty JSON body to the live stream endpoint. Prints the response body.

```go
package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.pyannote.ai/v1/live"

	payload := strings.NewReader("{}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}
```

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

### Main async function for live diarized transcription

Source: https://docs.pyannote.ai/tutorials/live-diarized-transcription-single-stt-stream

Sets up the event loop, creates a pyannoteAI stream, connects to OpenAI realtime API, configures transcription session, starts microphone, and runs tasks for diarization, transcription, and audio pumping. Handles graceful shutdown on Ctrl+C.

```python
async def main() -> None:
    loop = asyncio.get_running_loop()
    convo = Conversation()
    queue: "asyncio.Queue[np.ndarray]" = asyncio.Queue(maxsize=50)

    stream_id, url = create_stream()
    print(f"pyannoteAI stream: {stream_id}")

    client = AsyncOpenAI(api_key=OPENAI_API_KEY)
    async with websockets.connect(url) as ws:
        async with client.realtime.connect(extra_query={"intent": "transcription"}) as conn:
            await conn.session.update(session={
                "type": "transcription",
                "audio": {"input": {
                    "format": {"type": "audio/pcm", "rate": OPENAI_RATE},
                    "transcription": {"model": "gpt-realtime-whisper", "language": "en"},
                    "turn_detection": None,
                }},
            })

            mic = microphone(loop, queue)
            print("\nListening — speak now (Ctrl+C to stop)\n")

            stop = asyncio.Event()
            loop.add_signal_handler(signal.SIGINT, stop.set)
            tasks = [
                asyncio.create_task(pyannote_task(ws, convo, conn)),
                asyncio.create_task(openai_task(conn, convo)),
                asyncio.create_task(pump(queue, ws, convo, conn)),
            ]
            await asyncio.wait({*tasks, asyncio.create_task(stop.wait())}, return_when=asyncio.FIRST_COMPLETED)

            print("\nStopping...")
            mic.stop()
            mic.close()
            for task in tasks:
                task.cancel()
            await asyncio.gather(*tasks, return_exceptions=True)
            await ws.send(json.dumps({"type": "end_of_stream"}))


if __name__ == "__main__":
    asyncio.run(main())
```

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

### Example diarization output showing overlapping speech

Source: https://docs.pyannote.ai/features

Shows a JSON array of speaker segments with start and end times. Overlap occurs when segments from different speakers share time ranges, as seen between 12.5 and 14.0 seconds.

```json
[
  {
    "speaker": "SPEAKER_00",
    "start": 10.0,
    "end": 15.0
  },
  {
    "speaker": "SPEAKER_01",
    "start": 12.5,
    "end": 14.0
  }
]
```

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

### Go example for voiceprint extraction

Source: https://docs.pyannote.ai/api-reference/voiceprint

Uses net/http to POST the voiceprint payload and prints the response body.

```go
package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.pyannote.ai/v1/voiceprint"

	payload := strings.NewReader("{\n  \"url\": \"https://example.com/voice.wav\",\n  \"model\": \"precision-2\",\n  \"webhook\": \"https://example.com/webhook\",\n  \"webhookStatusOnly\": true\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}
```

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

### cURL: Get all jobs

Source: https://docs.pyannote.ai/api-reference/list-jobs

Sends a GET request to the /v2/jobs endpoint with an Authorization header.

```bash
curl --request GET \
  --url https://api.pyannote.ai/v2/jobs \
  --header 'Authorization: Bearer <token>'
```

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

### Upload audio file with curl

Source: https://docs.pyannote.ai/tutorials/how-to-upload-files

Uses curl to create a pre-signed PUT URL and then upload the file. The object-key is unique to your team. In production, set the API key as an environment variable.

```Shell
# Set your API key as an environment variable in production
# export PYANNOTE_API_KEY="your_api_key_here"

# Define your media object key
OBJECT_KEY="my-meeting-recording"

# Create pre-signed PUT URL
curl -X POST "https://api.pyannote.ai/v1/media/input" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"url\": \"media://${OBJECT_KEY}\"}"

# Upload file to the returned pre-signed URL
curl -X PUT "PRESIGNED_URL_FROM_RESPONSE" \
  -H "Content-Type: application/octet-stream" \
  --data-binary "@/path/to/your/audio.wav"
```

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

### Response 201: Upload URL

Source: https://docs.pyannote.ai/api-reference/upload-media

Example of a successful response with a pre-signed URL for uploading the media file.

```json
{
  "url": "<string>"
}
```

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

### Open N AssemblyAI streams

Source: https://docs.pyannote.ai/tutorials/live-diarized-transcription-per-speaker-stream

Creates one AssemblyAI WebSocket for each speaker slot before audio capture starts. Uses universal-3-5-pro with turn detection. AssemblyAI bills each open session independently, so N sessions produce N times the billed duration. Always terminate sessions when capture stops.

```python
params = urlencode({
    "sample_rate": SAMPLE_RATE,
    "encoding": "pcm_s16le",
    "speech_model": "universal-3-5-pro",
    "speaker_labels": "false",
    "min_turn_silence": 100,
    "max_turn_silence": 1000,
    "continuous_partials": "true",
})
assembly_url = f"wss://streaming.assemblyai.com/v3/ws?{params}"

aai_streams = [
    await stack.enter_async_context(
        websockets.connect(
            assembly_url,
            additional_headers={"Authorization": ASSEMBLYAI_API_KEY},
        )
    )
    for _ in range(max_speakers)
]
```

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

### PHP: Get all jobs

Source: https://docs.pyannote.ai/api-reference/list-jobs

Uses cURL in PHP to send a GET request and handle errors.

```php
<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.pyannote.ai/v2/jobs",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}
```

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

### Python example request with sensitivity parameters

Source: https://docs.pyannote.ai/tutorials/vad-crosstalk-sensitivity

Python script that sends a POST request to the diarize endpoint with vadSensitivity and crosstalkSensitivity set to 1.5. Replace YOUR_API_KEY with your actual API key.

```python
import requests

data = {
    "url": "https://files.pyannote.ai/marklex1min.wav",
    "model": "precision-3",
    "vadSensitivity": 1.5,
    "crosstalkSensitivity": 1.5,
}

response = requests.post(
    "https://api.pyannote.ai/v1/diarize",
    headers={"Authorization": "Bearer YOUR_API_KEY", "Content-Type": "application/json"},
    json=data,
)
print(response.json())
```

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

### JavaScript: Get all jobs

Source: https://docs.pyannote.ai/api-reference/list-jobs

Uses fetch to send a GET request and log the response or error.

```javascript
const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};

fetch('https://api.pyannote.ai/v2/jobs', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));
```

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

### JavaScript fetch example for /v1/identify

Source: https://docs.pyannote.ai/api-reference/identify

Uses fetch to POST the same payload to /v1/identify. Requires an authorization token and logs the JSON response or error.

```javascript
const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    voiceprints: [{label: 'John Doe', voiceprint: 'U29tZUJhc2U2NERhdGE'}],
    url: 'https://example.com/audio.wav',
    webhook: 'https://example.com/webhook',
    webhookStatusOnly: true,
    model: 'precision-2',
    numSpeakers: 2,
    minSpeakers: 1,
    maxSpeakers: 4,
    vadSensitivity: 0,
    crosstalkSensitivity: 0,
    turnLevelConfidence: true,
    speechProbability: true,
    crosstalkProbability: true,
    speakerProbability: true,
    exclusive: true,
    confidence: true,
    matching: {exclusive: true, threshold: 0}
  })
};

fetch('https://api.pyannote.ai/v1/identify', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));
```

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

### JavaScript example request with sensitivity parameters

Source: https://docs.pyannote.ai/tutorials/vad-crosstalk-sensitivity

JavaScript code using fetch to send a POST request to the diarize endpoint with vadSensitivity and crosstalkSensitivity set to 1.5. Replace YOUR_API_KEY with your actual API key.

```javascript
const data = {
  url: "https://files.pyannote.ai/marklex1min.wav",
  model: "precision-3",
  vadSensitivity: 1.5,
  crosstalkSensitivity: 1.5,
};

const response = await fetch("https://api.pyannote.ai/v1/diarize", {
  method: "POST",
  headers: { Authorization: "Bearer YOUR_API_KEY", "Content-Type": "application/json" },
  body: JSON.stringify(data),
});
console.log(await response.json());
```

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

### cURL GET request to test endpoint

Source: https://docs.pyannote.ai/api-reference/test

Sends a GET request to the test endpoint with Bearer token authorization.

```bash
curl --request GET \
  --url https://api.pyannote.ai/v1/test \
  --header 'Authorization: Bearer <token>'
```

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

### Upload audio file with TypeScript (axios)

Source: https://docs.pyannote.ai/tutorials/how-to-upload-files

Creates a pre-signed PUT URL via POST to /v1/media/input, then uploads a local file with PUT. The object-key is unique to your team. In production, use environment variables for the API key.

```TypeScript
import axios from 'axios';
import fs from 'fs';

async function uploadAudioFile(inputPath: string, objectKey: string, apiKey: string): Promise<void> {
  try {
    // Create the pre-signed PUT URL
    const response = await axios.post(
      'https://api.pyannote.ai/v1/media/input',
      { url: `media://${objectKey}` }, // Replace with your desired object-key
      {
        headers: {
          'Authorization': `Bearer ${apiKey}`, // In production, use environment variables: process.env.PYANNOTE_API_KEY
          'Content-Type': 'application/json'
        }
      }
    );

    const presignedUrl = response.data.url;

    // Upload local file to the pre-signed URL
    console.log(`Uploading ${inputPath} to ${presignedUrl}`);
    
    const fileData = fs.readFileSync(inputPath);
    await axios.put(presignedUrl, fileData, {
      headers: {
        'Content-Type': 'application/octet-stream'
      }
    });

    console.log('File uploaded successfully!');
  } catch (error) {
    console.error('Error uploading file:', error);
    throw error;
  }
}

// Usage
// uploadAudioFile('./path/to/your/audio.wav', 'my-meeting-recording', 'YOUR_API_KEY');
```
Tokens:
Raw