Actions12
Overview
This node integrates with the Coze AI platform, providing various AI-powered functionalities including audio transcription. Specifically, the Audio Transcription operation allows users to transcribe audio files into text by sending audio data to Coze's API.
Typical use cases include:
- Automatically converting recorded meetings, interviews, or podcasts into text for easier review and archiving.
- Transcribing voice notes or customer support calls to enable text-based analysis.
- Integrating transcription into workflows that require speech-to-text conversion as a step.
For example, you can feed an audio file from a previous node (like an HTTP request or file read node) into this node, specify the binary property containing the audio, and receive the transcription result in JSON format.
Properties
| Name | Meaning |
|---|---|
| Authentication | Method of authenticating with the Coze API. Options: - Service Token - OAuth2 |
| Binary Property | The name of the binary property on the input item that contains the audio file to transcribe |
Output
The node outputs a JSON array where each element corresponds to the transcription result of an input item. The exact structure depends on the Coze API response but generally includes:
- The transcribed text of the audio.
- Metadata related to the transcription (e.g., confidence scores, language detected).
- Any additional information returned by the API relevant to the transcription.
If the node supports binary data output, it would typically mean returning the original or processed audio file, but here the focus is on transcription text output.
Dependencies
- Requires access to the Coze AI platform via its public API at
https://api.coze.cn. - Needs either a service token or OAuth2 credentials configured in n8n for authentication.
- The node expects the input audio file to be provided as binary data under a specified property name.
Troubleshooting
Common issues:
- Incorrect or missing authentication credentials will cause authorization errors.
- Specifying a wrong binary property name will result in the node not finding the audio file to transcribe.
- Unsupported audio formats or corrupted files may lead to API errors or empty transcription results.
Error messages:
- Authorization failures: Check that the API key or OAuth2 token is valid and has necessary permissions.
- "Binary property not found": Verify the binary property name matches the input data.
- API request errors: Inspect the error message for details; ensure the audio file meets Coze API requirements.
Links and References
- Coze AI Platform Documentation (official API base URL)
- n8n documentation on working with binary data
- General info on audio transcription services