VeniceAi icon

VeniceAi

Use VeniceAi AI models.

Overview

The VeniceAi node's "Speech" operation converts input text into spoken audio using a selected synthetic voice. It supports various output audio formats and allows customization of the speech speed and streaming behavior. This node is useful for automating text-to-speech tasks, such as generating audio versions of messages, notifications, or content for accessibility purposes.

Practical examples:

  • Automatically generate audio files from customer support responses.
  • Create spoken alerts or announcements in workflow automations.
  • Convert blog posts or articles to audio for accessibility.

Properties

Name Type Meaning
Text string The text to convert to speech (max 4096 characters).
Voice options The synthetic voice to use for speech generation (various male/female voices available).
Response Format options The format of the generated audio file (MP3, Opus, AAC, FLAC, WAV, PCM).
Speed number The speed of the generated audio (range: 0.25 to 4.0; default is 1).
Streaming boolean If enabled, streams the audio back sentence by sentence.

Output

  • Binary Data:
    The node outputs the generated speech as binary audio data in the specified format (e.g., MP3, WAV, etc.).

    • The binary property name defaults to data unless otherwise configured.
    • The binary object includes metadata such as filename (speech.<format>) and MIME type (audio/<format>).
  • JSON Field:
    The json field is present but typically empty or contains only pass-through data from the input item.


Dependencies

  • External Service:
    Requires access to the Venice AI API.
  • API Key:
    You must configure valid Venice AI API credentials (veniceAiApi) in n8n, including:
    • apiKey: Your Venice AI API key.
    • baseUrl: Base URL for the Venice AI API (typically https://api.venice.ai).

Troubleshooting

Common Issues:

  • Missing API Key:
    Error: "No valid API key provided"
    Resolution: Ensure your Venice AI API credentials are correctly set up in n8n.

  • Invalid Input:

    • If the "Text" property is empty or exceeds 4096 characters, the request may fail.
    • Selecting an unsupported response format or voice may result in errors from the API.
  • Output Not Found:

    • If you do not see the expected binary output, check that the output property name matches your workflow configuration.

Error Handling:

  • The node will return an error message in the json.error field if an exception occurs and "Continue On Fail" is enabled.

Links and References

Discussion