Overview
The node "VLM Run" integrates with the VLM Run API to perform various media analysis and file management operations. Specifically, for the Analyze Audio operation, it allows users to upload audio files and analyze them for transcription, speaker identification, sentiment analysis, and other audio insights. This is useful in scenarios such as processing customer service calls, analyzing podcasts, or extracting meaningful data from any audio content.
Typical use cases include:
- Transcribing audio recordings automatically.
- Identifying speakers in multi-speaker audio.
- Performing sentiment analysis on spoken content.
- Integrating asynchronous processing workflows where results are delivered via a callback URL.
Properties
| Name | Meaning |
|---|---|
| File | The audio file data to analyze, typically coming from a previous node's output. |
| Domain Name or ID | The domain context to use for analysis; can be selected from a list or specified by an expression. |
| Process Asynchronously | Whether to process the audio analysis request asynchronously (true/false). |
| Callback URL | If asynchronous processing is enabled, this URL will be called when the analysis is complete. |
Output
The node outputs JSON data containing the results of the audio analysis. The structure depends on the response from the VLM Run API but generally includes transcription text, speaker labels, sentiment scores, and other metadata related to the audio content.
If asynchronous processing is enabled, the immediate output may contain information about the submitted job rather than final results, with the actual results delivered later to the provided callback URL.
No binary data is output by this node; all results are returned as structured JSON.
Dependencies
- Requires an API key credential for authenticating with the VLM Run API.
- Depends on the VLM Run API service being accessible.
- For asynchronous processing, requires a publicly accessible callback URL configured in the node.
- Uses internal helper utilities to handle file processing and uploading.
Troubleshooting
- File Upload Failures: Errors during file upload may occur if the input file data is missing or corrupted. Ensure the "File" property references valid binary data from a previous node.
- Invalid Domain: Selecting an invalid or empty domain may cause analysis requests to fail. Verify that the domain is correctly chosen or specified.
- Missing Callback URL: When asynchronous processing is enabled, omitting the callback URL will result in errors. Always provide a valid URL.
- API Errors: Network issues or invalid API credentials will cause the node to throw errors. Check API key validity and network connectivity.
- Multiple Items Handling: If processing multiple items, errors in one item can be isolated if "Continue On Fail" is enabled; otherwise, the node execution stops at the first error.
Common error messages:
"Failed to load domains: ..."— Indicates problems fetching available domains; check API access."The operation \"audio\" is not supported!"— Should not occur if the node is used correctly; indicates internal misconfiguration."At least one file mapping is required when multiple files is enabled"— Relevant for agent execution, not audio analysis.
Links and References
- n8n Expressions Documentation
- VLM Run API official documentation (not linked here due to lack of URL in source)