Wiro - Fish Audio S2 Pro TTS Model icon

Wiro - Fish Audio S2 Pro TTS Model

Fish Audio S2 Pro delivers fast, natural-sounding text-to-speech with voice cloning. Ideal for

Wiro - Fish Audio S2 Pro TTS Model

AI-generated

Overview

This node integrates with the Wiro Fish Audio S2 Pro TTS (Text-to-Speech) model to convert text prompts into natural-sounding speech audio. It supports advanced features like multi-speaker tags, emotion tags for prosody control, and voice cloning using reference audio and transcription. The node is useful for generating expressive and customizable speech outputs for applications such as voice assistants, audiobooks, or multimedia content.

Use Case Examples

  1. Generating speech from a text prompt with specific speaker and emotion tags.
  2. Cloning a voice by providing a reference audio URL and its exact transcription to produce personalized speech.
  3. Controlling speech generation randomness and length using parameters like temperature, max new tokens, and chunk length.

Properties

Name Meaning
Prompt Text to synthesize into speech; supports multi-speaker and emotion tags for prosody control.
Reference Audio URL URL of a reference audio file used for voice cloning; requires reference text to be set.
Reference Text Exact transcription of the reference audio file for accurate voice cloning.
Max New Tokens Maximum number of tokens to generate; 0 means automatic length based on input text.
Temperature Controls randomness in generation; lower values produce more consistent output, higher values increase expressiveness and variation.
Top P Nucleus sampling threshold; only tokens with cumulative probability up to this value are considered during generation.
Top K Limits token selection to the top K most probable tokens at each generation step.
Chunk Length Maximum UTF-8 byte length per generation chunk; smaller values may improve quality for long texts but increase processing time.
Seed Random seed for reproducibility; using the same seed with identical parameters produces the same audio output.

Output

JSON

  • taskid - Unique identifier for the TTS generation task.
  • url - URL to access the generated speech audio once the task is completed.
  • status - Status of the TTS generation task, e.g., 'completed' or 'failed'.

Dependencies

  • Requires Wiro API credentials including an API key and secret for authentication.

Troubleshooting

  • If the API response does not include a valid task ID or socket access token, the node throws an error indicating an invalid response from the Wiro API.
  • Task status codes '-1', '-2', '-3', and '-4' indicate failure states; users should verify input parameters and credentials if the task fails.
  • Network or authentication issues may prevent successful API requests; ensure valid credentials and network connectivity.

Discussion