Create Ambient Music Effortlessly with the MusicGen Cognitive Action

22 Apr 2025
Create Ambient Music Effortlessly with the MusicGen Cognitive Action

In today's fast-paced world, incorporating ambient music into applications can enhance user experiences significantly. The MusicGen Stereo Breathe Cognitive Actions provide developers with the capability to generate unique ambient music tracks with ease. By leveraging pre-built actions, developers can focus on integrating music generation into their applications without diving deep into complex audio processing algorithms.

Prerequisites

Before you start using the MusicGen Cognitive Actions, you need to have access to the platform, which typically requires an API key. This key will authenticate your requests to the service. Generally, you would pass the API key in the request headers to ensure secure access to the Cognitive Actions API.

Cognitive Actions Overview

Generate Ambient Music with MusicGen

The Generate Ambient Music with MusicGen action is designed to create ambient music tracks using the MusicGen Stereo Medium model. It utilizes a text token 'breathe' and offers various features such as seed and top-k sampling for variability, input audio for continuation or mimicry, and the ability to define multiple output formats.

Input

This action accepts the following input parameters:

  • seed (integer): Seed for the random number generator. If set to None or -1, a random seed will be generated.
  • topK (integer): Limits sampling to the k most probable tokens. Default is 250.
  • topP (number): Restricts sampling to tokens whose cumulative probability does not exceed p. Default is 0, meaning topK sampling is utilized.
  • prompt (string): A text prompt describing the style or characteristics of the music to be generated (e.g., "ambient, breathe").
  • duration (integer): Duration of the generated audio in seconds. Default is 8.
  • audioInput (string): URI of an audio file to influence the generated music.
  • temperature (number): Adjusts the variance of the sampling process. A higher value results in more diverse outputs. Default is 1.
  • continuation (boolean): Determines whether generated music will be a continuation of the provided melody. Default is false.
  • modelWeights (string): Specifies MusicGen model weights to utilize.
  • guidedInfluence (integer): Modifies the impact of input parameters on outputs. Default is 3.
  • audioOutputFormat (string): Defines the format for the generated audio output; either 'wav' or 'mp3'. Default is 'wav'.
  • audioContinuationEnd (integer): Specifies the end time for the audio file for continuation purposes.
  • useMultiBandDiffusion (boolean): Enables decoding EnCodec tokens using MultiBand Diffusion. Default is false.
  • audioContinuationStart (integer): Sets the start time for using the audio file in continuation. Default is 0.
  • audioNormalizationStrategy (string): Defines the method for audio normalization (e.g., 'loudness', 'clip', 'peak', 'rms'). Default is 'loudness'.

Example Input:

{
  "topK": 250,
  "topP": 0,
  "prompt": "ambient, breathe",
  "duration": 30,
  "temperature": 1,
  "continuation": false,
  "guidedInfluence": 3,
  "audioOutputFormat": "wav",
  "useMultiBandDiffusion": false,
  "audioContinuationStart": 0,
  "audioNormalizationStrategy": "loudness"
}

Output

The action typically returns a URL to the generated audio file. Here’s an example output:

https://assets.cognitiveactions.com/invocations/1a599e57-1a1b-4291-90eb-f11ca270224a/7a0f2448-8464-425a-b6b1-25b1749aa786.wav

Conceptual Usage Example (Python)

Below is a conceptual Python code snippet demonstrating how to call the MusicGen action:

import requests
import json

# Replace with your Cognitive Actions API key and endpoint
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute" # Hypothetical endpoint

action_id = "b543ede3-f4ca-487e-ba13-27b8f7727c6b" # Action ID for Generate Ambient Music with MusicGen

# Construct the input payload based on the action's requirements
payload = {
    "topK": 250,
    "topP": 0,
    "prompt": "ambient, breathe",
    "duration": 30,
    "temperature": 1,
    "continuation": false,
    "guidedInfluence": 3,
    "audioOutputFormat": "wav",
    "useMultiBandDiffusion": false,
    "audioContinuationStart": 0,
    "audioNormalizationStrategy": "loudness"
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json"
}

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json={"action_id": action_id, "inputs": payload} # Hypothetical structure
    )
    response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully:")
    print(json.dumps(result, indent=2))

except requests.exceptions.RequestException as e:
    print(f"Error executing action {action_id}: {e}")
    if e.response is not None:
        print(f"Response status: {e.response.status_code}")
        try:
            print(f"Response body: {e.response.json()}")
        except json.JSONDecodeError:
            print(f"Response body: {e.response.text}")

In this example, ensure to replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key. The action ID and input payload are structured according to the specifications of the MusicGen action. The endpoint URL is illustrative and may vary based on actual deployment.

Conclusion

The MusicGen Cognitive Action offers a robust and efficient way to generate ambient music tailored to your application's needs. With the capability to influence the generated audio through various parameters and inputs, developers can create unique soundscapes that enhance user engagement. Consider exploring different prompts and configurations to fully utilize this action's potential in your applications.

← Back to All Articles