Synthesizing Voices Effortlessly with Pheme Cognitive Actions

23 Apr 2025
Synthesizing Voices Effortlessly with Pheme Cognitive Actions

In the realm of text-to-speech capabilities, the Pheme Cognitive Actions provide a powerful tool for developers looking to integrate voice synthesis into their applications. With the ability to convert text into natural-sounding speech using a reference audio of a speaker's voice, these actions simplify the process of creating engaging audio content. This blog post will guide you through the key features and usage of the action available in the Pheme spec.

Prerequisites

Before diving into the integration of Pheme Cognitive Actions, ensure you have the following prerequisites:

  • API Key: You will need an API key to authenticate your requests to the Cognitive Actions platform. This key should be passed in the headers of your requests.
  • Speaker Audio Reference: Ensure you have a valid URI pointing to an audio file featuring the speaker's voice. This file will be essential for generating the synthesized speech.

Cognitive Actions Overview

Perform Voice Synthesis with Pheme

Description: This action allows you to utilize the Pheme model to synthesize voice from an input text and a speaker audio reference. It ensures that the content is meaningful and the speaker URI is valid.

Category: Text-to-Speech

Input

The input for this action is structured as follows:

  • Required Fields:
    • speakerReference: A URI pointing to an audio file of the speaker's voice. It must be accessible and correctly formatted.
    • text: The main body of text to be converted into speech, provided as a string. This is a required field and should contain meaningful content.
    • textReference: An optional supplementary string that elaborates on the main text, providing additional context.

Example Input:

{
  "text": "With tenure, Suzie'd have all the more leisure for yachting, but her publications are no good.",
  "textReference": "and keeping eternity before the eyes, though much.",
  "speakerReference": "https://replicate.delivery/pbxt/MNFXdPaUPOwYCZjZM4azsymbzE2TCV2WJXfGpeV2DrFWaSq8/example_en.wav"
}

Output

The output of this action is a URI to the synthesized audio file. Typically, this will return a link that you can use to access the generated speech file.

Example Output:

https://assets.cognitiveactions.com/invocations/5ea694d6-de99-4cdc-a300-a6242f2ea1cc/bc1d250e-a7f2-4cbb-85c6-f2abf4f55f9b.wav

Conceptual Usage Example (Python)

Here’s a conceptual Python code snippet demonstrating how you might call the Pheme voice synthesis 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 = "d17af11b-ceea-4631-b1a5-44f00081395b" # Action ID for Perform Voice Synthesis with Pheme

# Construct the input payload based on the action's requirements
payload = {
    "text": "With tenure, Suzie'd have all the more leisure for yachting, but her publications are no good.",
    "textReference": "and keeping eternity before the eyes, though much.",
    "speakerReference": "https://replicate.delivery/pbxt/MNFXdPaUPOwYCZjZM4azsymbzE2TCV2WJXfGpeV2DrFWaSq8/example_en.wav"
}

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 snippet, replace the placeholders with your actual API key and endpoint. The payload structure aligns with the required fields for the action, while the headers include the necessary authorization. The code handles potential errors gracefully, ensuring developers can debug effectively if something goes wrong.

Conclusion

The Pheme Cognitive Actions enable developers to seamlessly integrate voice synthesis capabilities into their applications, enhancing user engagement and accessibility. By leveraging the power of text-to-speech with personalized speaker references, you can create unique audio experiences. Explore further use cases, such as audio content creation, interactive voice applications, or enhancing gaming experiences, and start building today!

← Back to All Articles