AI - Transcribe
Type identifier:
ai:transcribeCategory: AI Operations
Description
The AI - Transcribe node converts spoken audio into text using OpenAI's Whisper model. It supports various audio formats and can handle multiple languages.
Input Handles
Handle | Type | Description |
|---|---|---|
|
| The audio file to transcribe. |
Output Handles
Handle | Type | Description |
|---|---|---|
|
| The transcribed text content. |
Configuration Options
Option | Type | Default | Description |
|---|---|---|---|
Model |
|
| The transcription model to use. |
See AI - Model Reference — Transcribe Models for the canonical model list.
Behaviour
- Receives audio file from input handle
- Transcribes the audio with the selected model
- Returns the transcribed text
Supported Audio Formats
- MP3
- MP4
- MPEG
- MPGA
- M4A
- WAV
- WEBM
Language Support
Whisper automatically detects the spoken language and provides transcription. It supports multiple languages with varying levels of accuracy.
File Size Limits
Audio files have size limits imposed by the API. Large recordings may need to be split.
Examples
Basic Audio Transcription
Input: Audio recording of a meeting
Output (transcription):
"Good morning everyone. Today we'll discuss the Q3 roadmap.
First, let's review the progress from last quarter..."Voice Message Processing
Flow:
- UX - Action receives audio from user
- AI - Transcribe converts to text
- AI - Chat Message processes the transcription
Multi-Language Support
Use case: Transcribe customer support calls in various languages, then translate or analyse the content.