Wiro - MOSS-TTSD
AI-generatedOverview
This node integrates with the Wiro MOSS-TTSD API to generate expressive multi-speaker conversational audio from a given dialogue text. It supports voice cloning by allowing up to five speakers to provide reference audio URLs and corresponding prompt texts. The node normalizes dialogue text optionally and handles asynchronous task polling to retrieve the final audio URL once processing is complete. This node is useful for creating realistic multi-speaker audio dialogues for applications such as voiceovers, audiobooks, or conversational AI.
Use Case Examples
- Generating a multi-speaker audio dialogue from a script with voice cloning for each speaker using their reference audio and prompt text.
- Normalizing and converting a dialogue text into expressive audio for use in multimedia projects.
Properties
| Name | Meaning |
|---|---|
| Dialogue | The text to generate audio from. |
| Speaker 1 Reference Audio URL | Reference audio URL for Speaker 1 used for voice cloning. Must be provided with Speaker 1 Text. Supported formats: .wav, .mp3, .m4a, .webm. |
| Speaker 1 Text | Prompt text for Speaker 1 used for voice cloning. Must be provided with Speaker 1 Reference Audio. The [S1] tag is added automatically if missing. |
| Speaker 2 Reference Audio URL | Reference audio URL for Speaker 2 used for voice cloning. Must be provided with Speaker 2 Text. Supported formats: .wav, .mp3, .m4a, .webm. |
| Speaker 2 Text | Prompt text for Speaker 2 used for voice cloning. Must be provided with Speaker 2 Reference Audio. The [S2] tag is added automatically if missing. |
| Speaker 3 Reference Audio URL | Reference audio URL for Speaker 3 used for voice cloning. Must be provided with Speaker 3 Text. Supported formats: .wav, .mp3, .m4a, .webm. |
| Speaker 3 Text | Prompt text for Speaker 3 used for voice cloning. Must be provided with Speaker 3 Reference Audio. The [S3] tag is added automatically if missing. |
| Speaker 4 Reference Audio URL | Reference audio URL for Speaker 4 used for voice cloning. Must be provided with Speaker 4 Text. Supported formats: .wav, .mp3, .m4a, .webm. |
| Speaker 4 Text | Prompt text for Speaker 4 used for voice cloning. Must be provided with Speaker 4 Reference Audio. The [S4] tag is added automatically if missing. |
| Speaker 5 Reference Audio URL | Reference audio URL for Speaker 5 used for voice cloning. Must be provided with Speaker 5 Text. Supported formats: .wav, .mp3, .m4a, .webm. |
| Speaker 5 Text | Prompt text for Speaker 5 used for voice cloning. Must be provided with Speaker 5 Reference Audio. The [S5] tag is added automatically if missing. |
| Text Normalize | Whether to enable normalization of dialogue text (removes special characters, fixes punctuation). Disabled by default. |
Output
JSON
taskid- The unique identifier of the audio generation task.url- The URL to the generated audio file once the task is completed.status- The status of the audio generation task, e.g., 'completed' or 'failed'.
Dependencies
- Wiro API key and secret credentials 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. Ensure the API credentials are correct and the request payload is valid.
- Task status codes '-1', '-2', '-3', '-4' indicate failure states; check the dialogue input and reference audio URLs for correctness and supported formats.
- Network or authentication errors may occur if the API key or secret is missing or invalid; verify credentials and network connectivity.
Links
- Wiro API Documentation - Official documentation for the Wiro API including MOSS-TTSD endpoint details.