text-to-speech

Transform Text into Captivating Audio with Audio LDM

25 Apr 2025
Transform Text into Captivating Audio with Audio LDM

In today's digital landscape, the ability to create engaging audio content from text is invaluable. With Audio LDM, developers can effortlessly generate audio assets, such as sound effects, human speech, and music, from simple text prompts. This powerful tool not only enhances the audio experience in applications but also streamlines content creation, saving time and resources. Whether you're building a game, developing an educational app, or creating audio stories, Audio LDM offers the flexibility and creativity needed to bring your ideas to life.

Common use cases for Audio LDM include generating immersive soundscapes for gaming, creating voiceovers for videos, and producing background music for various applications. By leveraging this service, developers can enhance user engagement and improve the overall experience.

Prerequisites

Before diving into Audio LDM, ensure you have a Cognitive Actions API key and a basic understanding of API call structures to effectively integrate these actions into your projects.

Generate Audio from Text with AudioLDM

The Generate Audio from Text with AudioLDM action allows you to create audio content from textual descriptions. This capability is particularly useful for developers looking to produce unique sound effects or voice simulations without the need for extensive audio libraries. The flexibility of this action supports advanced features like zero-shot text-guided audio style transfer, inpainting, and super-resolution, making it a versatile tool for audio generation.

Input Requirements

The action requires a CompositeRequest input structure, which includes:

  • text (string): The descriptive text prompt for the audio generation (e.g., "two starships are fighting in space with laser cannons").
  • duration (string): Specifies the length of the audio in seconds, with options ranging from 2.5 to 20.0 seconds (default is 5.0 seconds).
  • randomSeed (integer, optional): Influences the variability of the generated audio.
  • guidanceScale (number): Adjusts the model's focus on quality versus diversity, with a default value of 2.5.
  • numberOfCandidates (integer): Indicates how many audio variations to generate, with a default of 3.

Expected Output

The expected output is a URL link to the generated audio file, allowing easy access and integration into your applications. For example, an output might look like this: https://assets.cognitiveactions.com/invocations/f9e6e50a-04b4-431e-bab1-2e01da26a7f6/0b1bf580-3ceb-4ff8-8b39-110beb4b8e80.wav.

Use Cases for this Specific Action

This action is ideal for:

  • Game Development: Quickly generate sound effects for various in-game actions or environments.
  • Content Creation: Produce voiceovers for tutorials, podcasts, or storytelling applications, enhancing narrative delivery.
  • Interactive Media: Create dynamic audio responses based on user input, making applications feel more alive and engaging.
import requests
import json

# Replace with your actual Cognitive Actions API key and endpoint
# Ensure your environment securely handles the API key
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
# This endpoint URL is hypothetical and should be documented for users
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute"

action_id = "f70f0b8b-44f8-4034-bcbd-6609fa718e0b" # Action ID for: Generate Audio from Text with AudioLDM

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "text": "two starships are fighting in space with laser cannons",
  "duration": "5.0",
  "guidanceScale": 2.5,
  "numberOfCandidates": 3
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json",
    # Add any other required headers for the Cognitive Actions API
}

# Prepare the request body for the hypothetical execution endpoint
request_body = {
    "action_id": action_id,
    "inputs": payload
}

print(f"--- Calling Cognitive Action: {action.name or action_id} ---")
print(f"Endpoint: {COGNITIVE_ACTIONS_EXECUTE_URL}")
print(f"Action ID: {action_id}")
print("Payload being sent:")
print(json.dumps(request_body, indent=2))
print("------------------------------------------------")

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json=request_body
    )
    response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully. Result:")
    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 (non-JSON): {e.response.text}")
    print("------------------------------------------------")

Conclusion

With Audio LDM's ability to transform text into audio, developers can unlock new creative possibilities for their applications. Whether enhancing user experiences in gaming or generating engaging content for educational tools, the benefits of this service are clear. As you explore these actions, consider how they can be integrated into your projects to provide richer, more dynamic audio experiences. Embrace the power of Audio LDM and elevate your audio content creation to new heights.

← Back to All Articles