Actions74
- ai Actions
- array Actions
- audiovideo Actions
- carbone Actions
- charts Actions
- code Actions
- data Actions
- datetime Actions
- events Actions
- files Actions
- geo Actions
- hellospaces Actions
- images Actions
- Get Metadata of an Image
- Scale an Image
- Overlay an Image on another Image
- Scale an Image to Fit Given Dimensions
- Resize an Image
- Mask an Image with Another Image
- Mask an Image with a Circle Mask
- Crops an image by the given x, y, width and height paramters.
- Convert Texts to an Image
- Convert HTML String or URL to an Image
- Generates a QRCode with given text and color.
- Get EXIF from Image
- Remove EXIF from Image
- Modify Image EXIF
- playground Actions
- spreadsheet Actions
- stripe Actions
- youtube Actions
ai → Transcribe Audio with OpenAI
AI-generatedOverview
This node enables audio or video transcription using OpenAI's transcription capabilities via the 1001fx API. It is designed for scenarios where you need to convert spoken content in audio or video files into text, such as transcribing interviews, podcasts, meetings, lectures, or voice notes. The node supports both file uploads and remote file URLs, and allows customization of transcription parameters like language, output format, and model guidance.
Practical examples:
- Automatically transcribe customer support calls for analysis.
- Convert recorded webinars into searchable text documents.
- Generate subtitles from video files for accessibility.
Properties
| Name | Meaning |
|---|---|
| Audio file for transcription | The audio or video file to transcribe. You can upload a file directly. |
| File URL | The URL of an audio or video file to transcribe. Useful if your file is hosted online. |
| Prompt | Optional text to guide the model’s style or continue a previous audio segment. Should match the audio language. |
| Temperature | Sampling temperature (0–1). Higher values (e.g., 0.8) make output more random; lower values (e.g., 0.2) make it more focused and deterministic. Default is 0. |
| Language | The language of the input audio, in ISO-639-1 format (e.g., "en" for English, "es" for Spanish). Supplying this improves accuracy and latency. |
| Response Format | The format of the transcript output. Options: json (default), vtt. |
Output
If Response Format is set to
json, the output will be a JSON object containing the transcription result. The structure typically includes fields such as:text: The transcribed text.- Additional metadata may be present depending on the API response.
If Response Format is set to
vtt, the output will be a string in VTT (Web Video Text Tracks) subtitle format.The output is always available in the
jsonfield of the n8n item. If binary data is returned (not typical for this operation), it would be placed in thebinaryfield, but for transcription, expect only textual results.
Dependencies
- External Service: Requires access to the 1001fx API, which acts as a proxy to OpenAI's transcription service.
- API Key: The node requires the
1001fxApicredential to authenticate requests. - n8n Configuration: No special configuration beyond setting up the required credentials.
Troubleshooting
Common Issues:
- Invalid API Key: If the provided API key is incorrect or missing, the node will throw an authentication error. Ensure your
1001fxApicredentials are correctly configured. - Unsupported File Type: If the uploaded file is not a supported audio or video format, the API may return an error. Use standard formats like MP3, WAV, or MP4.
- Large Files: Very large files may cause timeouts or failures. Check the API documentation for file size limits.
- Language Code Errors: Providing an invalid or unsupported ISO-639-1 language code may reduce accuracy or cause errors.
Error Messages:
"error": "<message>"— The output will include anerrorfield with details if something goes wrong. Review the message for hints (e.g., authentication failure, file not found, etc.).- If the node is set to "Continue On Fail," errors will be included in the output; otherwise, the workflow will stop on error.