openrouter-transcribe

n8n node for transcribing audio to text using OpenRouter API

Package Information

Downloads: 276 weekly / 276 monthly
Latest Version: 0.1.2
Author: Boris Grit

Available Nodes

OpenRouter Transcribe icon
OpenRouter Transcribe
Transcribe audio to text using OpenRouter API

Documentation

n8n-nodes-openrouter-transcribe

This is an n8n community node that transcribes audio to text using the OpenRouter API.

Installation

Follow the installation guide to install this community node.

If you're using npm, run:

npm install n8n-nodes-openrouter-transcribe

If you're using pnpm, run:

pnpm add n8n-nodes-openrouter-transcribe

Usage

Credentials

  1. Get an API key from OpenRouter
  2. In n8n, create new credentials for OpenRouter API
  3. Enter your API key

Node Configuration

  1. Add the OpenRouter Transcribe node to your workflow
  2. Configure the following parameters:
    • Input Data: Choose whether to provide audio as binary data or base64 text
    • Binary Property: If using binary input, specify the property name containing the audio
    • Audio Format: Select the format of your audio file or choose "Auto-detect from filename" to automatically determine the format from the file extension
    • Model: Enter the model ID (default: openai/gpt-4o-audio-preview). Must be a model that supports audio input.
    • Prompt: Optional instruction to send along with the audio
    • Options: Additional parameters like max tokens, temperature, top P, and raw response option

Supported Audio Formats

  • WAV (.wav)
  • MP3 (.mp3)
  • AIFF (.aiff, .aif)
  • AAC (.aac)
  • OGG/OGA (.ogg, .oga)
  • FLAC (.flac)
  • M4A (.m4a)
  • PCM16 (.pcm16)
  • PCM24 (.pcm24)

When using "Auto-detect from filename", the format is determined by the file extension.

Example

  1. Use a Read Binary File node to load an audio file
  2. Connect it to the OpenRouter Transcribe node
  3. Set Binary Property to data (or whatever property your binary file uses)
  4. Set Audio Format to match your file format
  5. The output will contain the transcription text

Operations

The node returns:

  • transcription: The transcribed text
  • model: The model used for transcription
  • usage: Token usage information from the API

If Return Raw Response is enabled, the complete API response is returned instead.

Models

Only models with audio processing capabilities can transcribe audio. Find compatible models on the OpenRouter Models page.

Popular models include:

  • openai/gpt-4o-audio-preview
  • openai/whisper-large-v3

Check individual model documentation for supported audio formats and capabilities.

Resources

License

MIT

Discussion