Free Audio Transcribe

This node uses the free and open-source whisper-node package to transcribe audio files to text locally without requiring any API keys or external services. The node supports various audio formats and can automatically detect the language or use a specified language for transcription.

Features

• Completely Free: No API keys or subscriptions required
• Local Processing: All transcription happens on your machine
• Multiple Languages: Supports automatic language detection and specific language selection
• Various Audio Formats: Works with mp3, wav, m4a, and other common audio formats

Node Reference

Operations

• Transcribe: Transcribe an audio file to text

Configuration Fields

• Binary Property: The binary property that contains the audio file to transcribe
• Language: The language of the audio file (Auto-detect, English, German, Spanish, French, Italian, Japanese, Chinese)

Example Usage

1. Upload an audio file using the HTTP Request node or Read Binary File node
2. Connect the output to the Free Audio Transcribe node
3. Configure the node to use the correct binary property
4. Execute the workflow to get the transcription in the output

Technical Details

This node uses the whisper-node package, which is a Node.js wrapper around the Whisper speech recognition model. The first time you run the node, it will automatically download the necessary model files (approximately 75MB for the tiny model).

Supported Audio Formats

• mp3
• wav
• m4a
• flac
• ogg
• and more