Overview
This node integrates with Vocant AI to upload and process audio files for transcription. It allows users to send audio data to Vocant AI, check the status of transcription jobs, and retrieve the resulting transcriptions. This is useful in workflows where automated speech-to-text conversion is needed, such as generating captions, creating searchable archives of audio content, or automating note-taking from meetings.
Typical use cases include:
- Uploading an audio recording to Vocant AI for transcription.
- Polling Vocant AI to check if the transcription job is complete.
- Downloading the completed transcription text for further processing or storage.
Properties
| Name | Meaning |
|---|---|
| Audio File (binary) | The name(s) of the input binary field(s) containing the audio file(s) to be processed. Multiple fields can be specified separated by commas. |
| Callback URL | A URL where Vocant AI will send the results of the transcription asynchronously. Typically, this should be a webhook URL configured to receive transcription results automatically. |
| Use Original Filename | Whether to use the original filename of the uploaded audio file as the base name for the transcription file. If false, a default naming scheme is used. |
Output
The node outputs JSON data describing the result of each operation:
- Upload File: Outputs success status, job ID, file URL, expiration time, original filename, file size, and upload timestamp.
- Check Status: Outputs the current status information of the transcription job as returned by Vocant AI.
- Get Transcription: Outputs success status and job ID in JSON, and includes the transcription text as binary data encoded in base64 with MIME type
text/plain. The transcription file is named using the job ID.
If errors occur during processing, the output JSON contains an error field with the error message.
Dependencies
- Requires an API key credential for Vocant AI to authenticate requests.
- Makes HTTP requests to Vocant AI endpoints:
- To obtain a presigned upload URL.
- To upload the audio file.
- To check job status.
- To download transcription results.
- Optionally requires a publicly accessible callback URL to receive asynchronous transcription results.
Troubleshooting
- No binary data found for property "X": This error occurs if the specified binary input field does not exist or contains no data. Ensure the binary property name matches the actual input.
- HTTP request failures: Network issues or invalid API keys may cause request errors. Verify API credentials and network connectivity.
- Missing or incorrect callback URL: If using callbacks, ensure the URL is reachable and correctly configured to handle incoming data.
- Job ID errors: When checking status or retrieving transcription, ensure the correct job ID is provided, as returned by the upload operation.
To resolve these issues, verify input properties, credentials, and external endpoint availability.
Links and References
- Vocant AI Documentation (for API details and usage)
- n8n documentation on working with binary data
- General info on webhooks in n8n