Cloudflare AI

Overview

This node integrates with Cloudflare's AI services to perform various AI-powered transformations on text, images, audio, and embeddings. Specifically for the Audio - Speech to Text operation, it converts spoken audio content into written text using a speech recognition model.

Typical use cases include:

• Transcribing voice recordings or podcasts into text for further processing.
• Automating caption generation for videos.
• Extracting textual data from customer support calls or interviews.

For example, you can feed an audio file containing a meeting recording, and the node will output the transcribed text, enabling searchable archives or automated note-taking.

Properties

Output

The node outputs JSON data containing the transcription result under the json field. The structure typically includes:

• A top-level object with a result property holding the transcription text or related metadata returned by the Cloudflare Whisper model.
• If the transcription is successful, the success flag is implicitly true (no explicit flag shown).
• In case of failure, an error message is thrown or included if "Continue On Fail" is enabled.

No binary output is produced for this operation since the output is textual transcription.

Example output snippet (conceptual):

{
  "result": {
    "text": "Transcribed speech text here"
  }
}

Dependencies

• Requires a valid Cloudflare API key credential with access to the Cloudflare Workers AI models.
• The node uses the Cloudflare AI endpoint: https://api.cloudflare.com/client/v4/accounts/{accountId}/ai/run/@cf/openai/whisper
• The audio data must be provided as binary data within the specified binary property.
• Proper n8n credentials setup for the Cloudflare API key is necessary.

Troubleshooting

• Common issues:

  • Incorrect binary property name: Ensure the binary property name matches exactly the name of the binary field containing the audio.
  • Unsupported audio format or corrupted audio data may cause the request to fail.
  • Missing or invalid Cloudflare API credentials will prevent authentication.
  • Network or API rate limits could cause intermittent failures.

• Error messages:

  • "Request failed": Generic failure; check API token validity and network connectivity.
  • Errors including messages from the Cloudflare API are propagated; review the error details for specifics like invalid audio format or quota exceeded.

• Resolutions:

  • Verify the binary property name and ensure the audio file is correctly attached.
  • Confirm the Cloudflare API key is valid and has required permissions.
  • Check audio file format compatibility (commonly supported formats like WAV, MP3).
  • Enable "Continue On Fail" in the node settings to handle errors gracefully during batch processing.

Links and References

• Cloudflare Workers AI Documentation
• Whisper Speech-to-Text Model
• n8n Documentation on Binary Data Handling