### Install requests library

Source: https://github.com/omnisalegmbh/lalalai/blob/master/api-v1/python/README.md

Install the required Python library for making API requests.

```bash
pip install requests
```

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

### Start Stem Separation Task using Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Start a stem separation task using Python. This function allows customization of stems, extraction level, splitter, and optional lead/backing vocal separation. It returns the task_id for monitoring the process.

```python
import requests
import time

def split_stem(source_id: str, license_key: str, stem: str = "vocals",
               extraction_level: str = "deep_extraction",
               splitter: str = "auto", multivocal: str = None) -> str:
    """Start stem separation task. Returns task_id."""
    url = "https://www.lalal.ai/api/v1/split/stem_separator/"
    headers = {"X-License-Key": license_key}

    presets = {
        "stem": stem,  # vocals, drum, bass, piano, electric_guitar, acoustic_guitar, synthesizer, strings, wind
        "extraction_level": extraction_level,  # deep_extraction or clear_cut
        "splitter": splitter,  # auto, andromeda, perseus, orion, phoenix
    }
    if multivocal and stem == "vocals":
        presets["multivocal"] = "lead_back"  # Separate lead and backing vocals

    response = requests.post(url, json={"source_id": source_id, "presets": presets}, headers=headers)
    response.raise_for_status()
    return response.json()["task_id"]

# Extract drums from uploaded file
task_id = split_stem(source_id, "YOUR_LICENSE_KEY", stem="drum", splitter="andromeda")
print(f"Task started: {task_id}")

# Extract vocals with lead/backing separation
task_id = split_stem(source_id, "YOUR_LICENSE_KEY", stem="vocals", multivocal="lead_back")
```

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

### Start Stem Separation Task

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Initiate a stem separation task for a given source_id. This API endpoint allows specifying the desired stem (e.g., vocals, drums), extraction level, and the neural network splitter to use. The response contains a task_id for polling.

```bash
curl -X POST "https://www.lalal.ai/api/v1/split/stem_separator/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "source_id": "abc123-source-id",
    "presets": {
      "stem": "vocals",
      "extraction_level": "deep_extraction",
      "splitter": "andromeda",
      "dereverb_enabled": false
    }
  }'
```

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

### Run Voice Conversion with lalalai_voice_converter.py

Source: https://github.com/omnisalegmbh/lalalai/blob/master/api-v1/python/README.md

Use this script for voice conversion tasks. It requires a user license and allows specifying input/output paths, voice pack, accent, tonality, dereverb settings, and splitter model. Use `--list` to see available voice packs.

```bash
python3 lalalai_voice_converter.py --license <user license> \
                                   [--input <input directory or file>] \
                                   [--uploaded_file_id <file id>] \
                                   [--output <output directory>] \
                                   [--voice_pack_id <voice pack id>] \
                                   [--accent <0.0-1.0>] \
                                   [--tonality-reference <source_file/voice_pack>] \
                                   [--dereverb-enabled <true/false>] \
                                   [--splitter <model>] \
                                   [--delete <true/false>] \
                                   [--list]
```

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

### Run lalalai_splitter.py script

Source: https://github.com/omnisalegmbh/lalalai/blob/master/api-v1/python/README.md

Execute the audio source separation script with various command-line arguments. Specify license, input, and optional output, stem, noise cancelling, dereverb, extraction level, multivocal, splitter model, and delete options.

```bash
python3 lalalai_splitter.py --license <user license> \
                            --input <input directory or file> \
                            [--output <output directory>] \
                            [--stem <stem option>] \
                            [--noise-cancelling <level>] \
                            [--dereverb-enabled <true/false>] \
                            [--extraction-level <level>] \
                            [--multivocal <option>] \
                            [--splitter <model>] \
                            [--delete <true/false>]
```

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

### Upload File to LALAL.AI

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Upload an audio or video file to LALAL.AI storage. The file is sent as raw binary data with the filename in the Content-Disposition header. Requires authentication with an X-License-Key.

```bash
curl -X POST "https://www.lalal.ai/api/v1/upload/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Disposition: attachment; filename=\"song.mp3\"" \
  --data-binary @song.mp3
```

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

### List Available Voice Packs with Bash

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Use this curl command to retrieve a list of available AI voice packs. Requires your license key.

```bash
# List available voice packs
curl -X POST "https://www.lalal.ai/api/v1/voice_packs/list/" \
  -H "X-License-Key: YOUR_LICENSE_KEY"
```

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

### List Available Voice Packs with Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

This Python function fetches a list of all available AI voice packs that are ready for use. It requires your license key for authentication.

```python
def list_voice_packs(license_key: str) -> list:
    """Get available voice packs."""
    url = "https://www.lalal.ai/api/v1/voice_packs/list/"
    response = requests.post(url, headers={"X-License-Key": license_key})
    response.raise_for_status()
    return [p for p in response.json()["packs"] if p["ready_to_use"]]
```

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

### Upload File using Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Upload a file to LALAL.AI using Python's requests library. This function returns the source_id needed for subsequent processing tasks. Ensure the file path and license key are correct.

```python
import requests

def upload_file(file_path: str, license_key: str) -> str:
    """Upload file to LALAL.AI and return source_id."""
    url = "https://www.lalal.ai/api/v1/upload/"
    filename = file_path.split("/")[-1]
    headers = {
        "Content-Disposition": f'attachment; filename="{filename}"',
        "X-License-Key": license_key,
    }
    with open(file_path, "rb") as f:
        response = requests.post(url, data=f, headers=headers, timeout=30)
    response.raise_for_status()
    return response.json()["id"]  # Returns source_id for processing

# Usage
source_id = upload_file("my_song.mp3", "YOUR_LICENSE_KEY")
print(f"Uploaded file with source_id: {source_id}")
```

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

### Upload File

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Upload an audio or video file to LALAL.AI storage and receive a `source_id` for subsequent processing operations. The file is sent as raw binary data in the request body with filename specified in the Content-Disposition header.

```APIDOC
## POST /api/v1/upload/

### Description
Upload an audio or video file to LALAL.AI storage and receive a `source_id` for subsequent processing operations.

### Method
POST

### Endpoint
/api/v1/upload/

### Parameters
#### Headers
- **X-License-Key** (string) - Required - Your license key for authentication.
- **Content-Disposition** (string) - Required - Specifies the filename of the uploaded file (e.g., `attachment; filename="song.mp3"`).

#### Request Body
- **file** (binary) - Required - The raw binary data of the audio or video file.

### Request Example
```bash
curl -X POST "https://www.lalal.ai/api/v1/upload/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Disposition: attachment; filename=\"song.mp3\"" \
  --data-binary @song.mp3
```

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the uploaded source file (`source_id`).
- **filename** (string) - The name of the uploaded file.
- **duration** (float) - The duration of the audio/video file in seconds.

#### Response Example
```json
{
  "id": "abc123-source-id",
  "filename": "song.mp3",
  "duration": 180.5
}
```
```

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

### Delete Files with Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Deletes source files and all resulting tracks from LALAL.AI storage. Requires the source ID and license key.

```python
import requests

def delete_files(source_id: str, license_key: str):
    """Delete source file and all resulting tracks from LALAL.AI storage."""
    url = "https://www.lalal.ai/api/v1/delete/"
    headers = {"X-License-Key": license_key}

    response = requests.post(url, json={"source_id": source_id}, headers=headers)
    response.raise_for_status()
    print(f"Deleted files for source_id: {source_id}")

# Clean up after downloading results
# delete_files(source_id, "YOUR_LICENSE_KEY")
```

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

### Download Track with Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Downloads an audio track from a given URL and saves it to a specified directory. It attempts to extract the filename from the Content-Disposition header, falling back to the URL's last segment.

```python
import requests

def download_track(url: str, output_dir: str) -> str:
    """Download track from URL and save to output directory."""
    response = requests.get(url, stream=True, timeout=60)
    response.raise_for_status()

    # Extract filename from Content-Disposition header
    content_disp = response.headers.get("Content-Disposition", "")
    if "filename=" in content_disp:
        filename = content_disp.split("filename=")[1].strip('"')
    else:
        filename = url.split("/")[-1]

    file_path = f"{output_dir}/{filename}"
    with open(file_path, "wb") as f:
        for chunk in response.iter_content(chunk_size=8192):
            f.write(chunk)

    return file_path

# Download all tracks from result
# for track in result["tracks"]:
#     file_path = download_track(track["url"], "./output")
#     print(f"Downloaded {track['label']} to {file_path}")
```

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

### Delete Files with cURL

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Deletes source and processed files from LALAL.AI storage using a POST request. Requires a license key and the source ID.

```bash
# Delete source and result files
curl -X POST "https://www.lalal.ai/api/v1/delete/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '{"source_id": "abc123-source-id"}'

# Response:
# {"status": "ok"}
```

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

### Convert Voice with AI Voice Pack using Bash

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

This bash command demonstrates how to change a voice using a specified AI voice pack. You can adjust accent and tonality, and control dereverberation.

```bash
# Convert voice using a voice pack
curl -X POST "https://www.lalal.ai/api/v1/change_voice/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "source_id": "abc123-source-id",
    "presets": {
      "voice_pack_id": "ALEX_KAYE",
      "accent": 1.0,
      "tonality_reference": "voice_pack",
      "dereverb_enabled": false
    }
  }'
```

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

### Python Class for LALAL.AI API v1 Workflow

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

This class encapsulates the complete API workflow: uploading files, initiating processing tasks (like splitting vocals), polling for results, downloading processed tracks, and cleaning up resources. It requires a license key for authentication and uses the 'requests' library for HTTP communication.

```python
import requests
import time
import os

class LalalAI:
    BASE_URL = "https://www.lalal.ai/api/v1/"

    def __init__(self, license_key: str):
        self.license_key = license_key
        self.headers = {"X-License-Key": license_key}

    def upload(self, file_path: str) -> str:
        filename = os.path.basename(file_path)
        headers = {**self.headers, "Content-Disposition": f'attachment; filename="{filename}"}
        with open(file_path, "rb") as f:
            resp = requests.post(f"{self.BASE_URL}upload/", data=f, headers=headers)
        resp.raise_for_status()
        return resp.json()["id"]

    def split_vocals(self, source_id: str, splitter: str = "andromeda") -> str:
        resp = requests.post(f"{self.BASE_URL}split/stem_separator/",
            json={"source_id": source_id, "presets": {"stem": "vocals", "extraction_level": "deep_extraction", "splitter": splitter}},
            headers=self.headers)
        resp.raise_for_status()
        return resp.json()["task_id"]

    def wait_for_result(self, task_id: str) -> dict:
        while True:
            resp = requests.post(f"{self.BASE_URL}check/", json={"task_ids": [task_id]}, headers=self.headers)
            resp.raise_for_status()
            task = resp.json()["result"][task_id]
            if task["status"] == "success":
                return task["result"]
            elif task["status"] == "cancelled":
                raise RuntimeError("Task cancelled")
            print(f"Progress: {task.get('progress', 0)}%")
            time.sleep(5)

    def download(self, url: str, output_dir: str) -> str:
        resp = requests.get(url, stream=True)
        filename = resp.headers.get("Content-Disposition", "").split("filename=")[-1].strip('"') or "output.mp3"
        path = os.path.join(output_dir, filename)
        with open(path, "wb") as f:
            for chunk in resp.iter_content(8192):
                f.write(chunk)
        return path

    def delete(self, source_id: str):
        requests.post(f"{self.BASE_URL}delete/", json={"source_id": source_id}, headers=self.headers)

# Usage: Extract vocals from a song
api = LalalAI("YOUR_LICENSE_KEY")
source_id = api.upload("song.mp3")
task_id = api.split_vocals(source_id)
result = api.wait_for_result(task_id)

for track in result["tracks"]:
    path = api.download(track["url"], "./output")
    print(f"Downloaded {track['label']}: {path}")

api.delete(source_id)  # Clean up

```

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

### Perform Music Separation with lalalai_demuser.py

Source: https://github.com/omnisalegmbh/lalalai/blob/master/api-v1/python/README.md

This script is used for separating background music from voice. It requires a user license and input files, with optional parameters for output directory, dereverb, and deletion of files from storage.

```bash
python3 lalalai_demuser.py --license <user license> \
                           --input <input directory or file> \
                           [--output <output directory>] \
                           [--dereverb-enabled <true/false>] \
                           [--delete <true/false>]
```

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

### POST /api/v1/split/demuser/

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Removes background music from voice recordings, leaving clean speech. Ideal for extracting vocals from videos with background music or cleaning up interviews.

```APIDOC
## POST /api/v1/split/demuser/

### Description
Removes background music from voice recordings, leaving clean speech. Ideal for extracting vocals from videos with background music or cleaning up interviews.

### Method
POST

### Endpoint
https://www.lalal.ai/api/v1/split/demuser/

### Parameters
#### Request Body
- **source_id** (string) - Required - The ID of the source audio file.
- **presets** (object) - Required - Configuration for the demuser process.
  - **stem** (string) - Required - Should be set to "music".
  - **dereverb_enabled** (boolean) - Optional - Whether to enable dereverberation. Defaults to false.

### Request Example
```json
{
  "source_id": "abc123-source-id",
  "presets": {
    "stem": "music",
    "dereverb_enabled": false
  }
}
```

### Response
#### Success Response (200)
- **task_id** (string) - The ID of the processing task.

#### Response Example
```json
{
  "task_id": "task-demuser-123"
}
```
```

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

### Stem Separator - Extract Instruments and Vocals

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Separate audio into individual stems including vocals, drums, bass, piano, guitars, synthesizer, strings, and wind instruments. Supports multiple neural networks and extraction modes for quality control.

```APIDOC
## POST /api/v1/split/stem_separator/

### Description
Initiates a stem separation task to extract individual audio components like vocals, drums, bass, etc., from a source file.

### Method
POST

### Endpoint
/api/v1/split/stem_separator/

### Parameters
#### Headers
- **X-License-Key** (string) - Required - Your license key for authentication.

#### Request Body
- **source_id** (string) - Required - The ID of the source file obtained from the upload endpoint.
- **presets** (object) - Required - Configuration for the stem separation process.
  - **stem** (string) - Required - The type of stem to extract (e.g., `vocals`, `drum`, `bass`, `piano`, `electric_guitar`, `acoustic_guitar`, `synthesizer`, `strings`, `wind`).
  - **extraction_level** (string) - Optional - Specifies the extraction quality (`deep_extraction` or `clear_cut`). Defaults to `deep_extraction`.
  - **splitter** (string) - Optional - The neural network model to use for separation (`auto`, `andromeda`, `perseus`, `orion`, `phoenix`). Defaults to `auto`.
  - **dereverb_enabled** (boolean) - Optional - Whether to enable de-reverberation. Defaults to `false`.
  - **multivocal** (string) - Optional - If `stem` is `vocals`, this can be `lead_back` to separate lead and backing vocals.

### Request Example
```bash
curl -X POST "https://www.lalal.ai/api/v1/split/stem_separator/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '{ 
    "source_id": "abc123-source-id",
    "presets": {
      "stem": "vocals",
      "extraction_level": "deep_extraction",
      "splitter": "andromeda",
      "dereverb_enabled": false
    }
  }'
```

### Response
#### Success Response (200)
- **task_id** (string) - The unique identifier for the initiated processing task.

#### Response Example
```json
{
  "task_id": "task-xyz-789"
}
```
```

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

### Poll Task Status with Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Polls the LALAL.AI API to check the status of a processing task. Requires a license key and task ID. Returns download URLs upon successful completion.

```python
import time
import requests

def wait_for_task(task_id: str, license_key: str, poll_interval: int = 5) -> dict:
    """Poll task until completion. Returns result with download URLs."""
    url = "https://www.lalal.ai/api/v1/check/"
    headers = {"X-License-Key": license_key}

    while True:
        response = requests.post(url, json={"task_ids": [task_id]}, headers=headers)
        response.raise_for_status()

        task_status = response.json()["result"][task_id]
        status = task_status.get("status")

        if status == "success":
            print("Processing complete!")
            return task_status["result"]
        elif status == "cancelled":
            raise RuntimeError("Task was cancelled")
        elif status == "progress":
            progress = task_status.get("progress", 0)
            print(f"Progress: {progress}%")

        time.sleep(poll_interval)

# Wait for task and get download URLs
# result = wait_for_task(task_id, "YOUR_LICENSE_KEY")
# for track in result["tracks"]:
#     print(f"{track['label']}: {track['url']}")
```

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

### Download Processed Audio with cURL

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Downloads a processed audio file using a direct URL obtained from the task check response. The `-J` flag attempts to use the filename from the Content-Disposition header.

```bash
# Download result file (URL from check response)
curl -O -J "https://www.lalal.ai/api/v1/download/abc123/vocals.mp3"
```

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

### Clean Voice Recording with Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Use this function to clean voice recordings by adjusting noise levels and optionally removing reverb. Requires a source ID and license key.

```python
def voice_clean(source_id: str, license_key: str,
                noise_level: int = 1, dereverb: bool = False) -> str:
    """Clean voice recording. noise_level: 0=mild, 1=normal, 2=aggressive."""
    url = "https://www.lalal.ai/api/v1/split/voice_clean/"
    headers = {"X-License-Key": license_key}

    presets = {
        "stem": "voice",
        "noise_cancelling_level": noise_level,
        "splitter": "lynx",  # Fastest for voice cleaning
    }
    if dereverb:
        presets["dereverb_enabled"] = True

    response = requests.post(url, json={"source_id": source_id, "presets": presets}, headers=headers)
    response.raise_for_status()
    return response.json()["task_id"]
```

```python
# Clean a podcast recording with echo removal
task_id = voice_clean(source_id, "YOUR_LICENSE_KEY", noise_level=2, dereverb=True)
```

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

### Change Voice using AI Voice Pack with Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Use this Python function to transform voice recordings with AI voice packs. Customize accent and tonality, and optionally enable dereverb. Requires source ID, license key, and voice pack ID.

```python
def change_voice(source_id: str, license_key: str, voice_pack_id: str = "ALEX_KAYE",
                 accent: float = 1.0, tonality: str = "source_file") -> str:
    """Convert voice using AI voice pack.
    accent: 0.0=keep original, 1.0=match voice pack
    tonality: 'source_file'=keep pitch, 'voice_pack'=match voice pack pitch
    """
    url = "https://www.lalal.ai/api/v1/change_voice/"
    headers = {"X-License-Key": license_key}

    presets = {
        "voice_pack_id": voice_pack_id,
        "accent": accent,
        "tonality_reference": tonality,
        "dereverb_enabled": False,
    }

    response = requests.post(url, json={"source_id": source_id, "presets": presets}, headers=headers)
    response.raise_for_status()
    return response.json()["task_id"]
```

```python
# List and use voice packs
packs = list_voice_packs("YOUR_LICENSE_KEY")
for pack in packs[:5]:
    print(f"{pack['pack_id']}: {pack['name']}")

task_id = change_voice(source_id, "YOUR_LICENSE_KEY", voice_pack_id="ALEX_KAYE", accent=0.8)
```

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

### POST /api/v1/check/

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Polls the status of processing tasks. Tasks can be in 'progress' with percentage updates, or reach 'success' or 'cancelled' states.

```APIDOC
## POST /api/v1/check/

### Description
Polls the status of processing tasks. Tasks can be in 'progress' with percentage updates, or reach 'success' or 'cancelled' states.

### Method
POST

### Endpoint
https://www.lalal.ai/api/v1/check/

### Parameters
#### Request Body
- **task_ids** (array of strings) - Required - A list of task IDs to check the status for.

### Request Example
```json
{
  "task_ids": ["task-xyz-789", "task-voice-clean-456"]
}
```

### Response
#### Success Response (200)
- **result** (object) - An object where keys are task IDs and values are status objects.
  - **status** (string) - The current status of the task (e.g., "progress", "success", "cancelled").
  - **progress** (integer) - The completion percentage of the task (if status is "progress").
  - **presets** (object) - The presets used for the task (if available).

#### Response Example (in progress)
```json
{
  "result": {
    "task-xyz-789": {"status": "progress", "progress": 45, "presets": {...}}
  }
}
```

#### Response Example (completed)
```json
{
  "result": {
    "task-xyz-789": {"status": "success", "presets": {...}}
  }
}
```
```

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

### Clean Voice with Noise Cancellation

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Initiate a voice cleaning task to remove background noise and optionally echo/reverb. This endpoint allows specifying the noise cancellation level and the splitter to use. Requires a source_id and authentication.

```bash
curl -X POST "https://www.lalal.ai/api/v1/split/voice_clean/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "source_id": "abc123-source-id",
    "presets": {
      "stem": "voice",
      "noise_cancelling_level": 2,
      "dereverb_enabled": true,
      "splitter": "lynx"
    }
  }'
```

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

### Check Task Status with Bash

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

This bash command polls the status of a processing task. It requires a list of task IDs and your license key. The response indicates the status and progress.

```bash
# Check task status
curl -X POST "https://www.lalal.ai/api/v1/check/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '{"task_ids": ["task-xyz-789"]}'
```

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

### Remove Background Music with Bash

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Use this curl command to remove background music from an audio source. Specify the source ID and license key. Set `dereverb_enabled` to true if needed.

```bash
# Remove background music from audio
curl -X POST "https://www.lalal.ai/api/v1/split/demuser/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "source_id": "abc123-source-id",
    "presets": {
      "stem": "music",
      "dereverb_enabled": false
    }
  }'
```

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

### POST /api/v1/split/voice_clean/

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Cleans voice recordings by removing noise and optionally reverb. Useful for improving the quality of speech.

```APIDOC
## POST /api/v1/split/voice_clean/

### Description
Cleans voice recordings by removing noise and optionally reverb. Useful for improving the quality of speech.

### Method
POST

### Endpoint
https://www.lalal.ai/api/v1/split/voice_clean/

### Parameters
#### Request Body
- **source_id** (string) - Required - The ID of the source audio file.
- **presets** (object) - Required - Configuration for the voice cleaning process.
  - **stem** (string) - Required - Should be set to "voice".
  - **noise_cancelling_level** (integer) - Optional - Level of noise cancellation. 0=mild, 1=normal, 2=aggressive. Defaults to 1.
  - **splitter** (string) - Optional - The splitter algorithm to use. Defaults to "lynx".
  - **dereverb_enabled** (boolean) - Optional - Whether to enable dereverberation. Defaults to false.

### Request Example
```json
{
  "source_id": "abc123-source-id",
  "presets": {
    "stem": "voice",
    "noise_cancelling_level": 2,
    "dereverb_enabled": true
  }
}
```

### Response
#### Success Response (200)
- **task_id** (string) - The ID of the processing task.

#### Response Example
```json
{
  "task_id": "task-voice-clean-456"
}
```
```

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

### Voice Changer API

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Transforms voice recordings using AI voice packs. Allows adjustment of accent and tonality, with an option to remove reverb.

```APIDOC
## POST /api/v1/voice_packs/list/

### Description
Retrieves a list of available AI voice packs.

### Method
POST

### Endpoint
https://www.lalal.ai/api/v1/voice_packs/list/

### Response
#### Success Response (200)
- **packs** (array) - A list of available voice packs. Each pack is an object with:
  - **pack_id** (string) - The unique identifier for the voice pack.
  - **name** (string) - The display name of the voice pack.
  - **ready_to_use** (boolean) - Indicates if the voice pack is ready for use.

#### Response Example
```json
{
  "packs": [
    {"pack_id": "ALEX_KAYE", "name": "Alex Kaye", "ready_to_use": true},
    ...
  ]
}
```

## POST /api/v1/change_voice/

### Description
Converts a voice recording using a specified AI voice pack. Allows customization of accent and tonality.

### Method
POST

### Endpoint
https://www.lalal.ai/api/v1/change_voice/

### Parameters
#### Request Body
- **source_id** (string) - Required - The ID of the source audio file.
- **presets** (object) - Required - Configuration for the voice changing process.
  - **voice_pack_id** (string) - Required - The ID of the voice pack to use.
  - **accent** (float) - Optional - Controls the accent strength. 0.0 means keep original accent, 1.0 means match the voice pack's accent. Defaults to 1.0.
  - **tonality_reference** (string) - Optional - Determines how tonality is handled. Options are 'source_file' (keep original pitch) or 'voice_pack' (match voice pack pitch). Defaults to 'source_file'.
  - **dereverb_enabled** (boolean) - Optional - Whether to enable dereverberation. Defaults to false.

### Request Example
```json
{
  "source_id": "abc123-source-id",
  "presets": {
    "voice_pack_id": "ALEX_KAYE",
    "accent": 0.8,
    "tonality_reference": "voice_pack",
    "dereverb_enabled": false
  }
}
```

### Response
#### Success Response (200)
- **task_id** (string) - The ID of the processing task.

#### Response Example
```json
{
  "task_id": "task-change-voice-789"
}
```
```

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

### Remove Background Music with Python

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

This Python function removes background music from voice recordings. It takes a source ID and license key, with an option to enable dereverberation.

```python
def demuser(source_id: str, license_key: str, dereverb: bool = False) -> str:
    """Remove background music from voice recording."""
    url = "https://www.lalal.ai/api/v1/split/demuser/"
    headers = {"X-License-Key": license_key}

    presets = {"stem": "music"}
    if dereverb:
        presets["dereverb_enabled"] = True

    response = requests.post(url, json={"source_id": source_id, "presets": presets}, headers=headers)
    response.raise_for_status()
    return response.json()["task_id"]
```

```python
# Remove background music from a video interview
task_id = demuser(source_id, "YOUR_LICENSE_KEY", dereverb=True)
```

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

### Voice Clean - Remove Background Noise

Source: https://context7.com/omnisalegmbh/lalalai/llms.txt

Clean voice recordings by removing background noise, hum, and unwanted sounds. Supports adjustable noise cancellation levels and optional echo/reverb removal.

```APIDOC
## POST /api/v1/split/voice_clean/

### Description
Initiates a voice cleaning task to remove background noise, hum, and echo/reverb from voice recordings.

### Method
POST

### Endpoint
/api/v1/split/voice_clean/

### Parameters
#### Headers
- **X-License-Key** (string) - Required - Your license key for authentication.

#### Request Body
- **source_id** (string) - Required - The ID of the source file obtained from the upload endpoint.
- **presets** (object) - Required - Configuration for the voice cleaning process.
  - **stem** (string) - Required - Should be set to `voice` for voice cleaning.
  - **noise_cancelling_level** (integer) - Optional - The intensity of noise cancellation. Higher values mean more aggressive cancellation. Defaults to `1`.
  - **dereverb_enabled** (boolean) - Optional - Whether to enable de-reverberation. Defaults to `false`.
  - **splitter** (string) - Optional - The neural network model to use for processing (e.g., `lynx`). Defaults to `auto`.

### Request Example
```bash
curl -X POST "https://www.lalal.ai/api/v1/split/voice_clean/" \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '{ 
    "source_id": "abc123-source-id",
    "presets": {
      "stem": "voice",
      "noise_cancelling_level": 2,
      "dereverb_enabled": true,
      "splitter": "lynx"
    }
  }'
```

### Response
#### Success Response (200)
- **task_id** (string) - The unique identifier for the initiated processing task.

#### Response Example
```json
{
  "task_id": "task-abc-123"
}
```
```

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