Overview
This node streams audio from a microphone device to AssemblyAI's real-time transcription service. It establishes a WebSocket connection with AssemblyAI, sends live audio data captured from the specified microphone, and receives partial and final transcriptions in real time. This is useful for scenarios such as live captioning, voice command recognition, or any application requiring immediate speech-to-text conversion.
For example, you can use this node to transcribe a meeting or lecture as it happens, enabling instant text output that can be logged, analyzed, or displayed.
Properties
| Name | Meaning |
|---|---|
| Duration (seconds) | Duration to record audio in seconds. Set to 0 for continuous streaming until manually stopped. |
| Device | Microphone device to use for audio input. Use "default" to select the system's default microphone. |
Output
The node outputs an array of objects representing transcription messages received during the streaming session. Each object has a json field containing:
text: The transcribed text segment.timestamp: ISO string timestamp when the transcription was received.confidence: (only for final transcripts) Confidence score of the transcription.message_type: Either"final"for finalized transcriptions or"partial"for interim results.
The output does not include binary data; it focuses on textual transcription results streamed from AssemblyAI.
Dependencies
- Requires an active API key credential for AssemblyAI to authenticate requests.
- Uses the WebSocket protocol to connect to AssemblyAI's real-time transcription endpoint.
- Depends on a microphone input device accessible by the host system.
- Requires n8n environment to have access to the microphone hardware and permissions to capture audio.
Troubleshooting
- No credentials provided!: Ensure that a valid AssemblyAI API key credential is configured in n8n.
- Failed to get AssemblyAI token: Check network connectivity and validity of the API key. The error message includes HTTP status and response details.
- Invalid response from AssemblyAI: No token received: Indicates an unexpected response from the token endpoint; verify API key and AssemblyAI service status.
- Microphone error: Could be caused by lack of permission to access the microphone, incorrect device name, or hardware issues.
- WebSocket error: Network interruptions or invalid WebSocket URL/token can cause this. Verify internet connection and token validity.
- If the node stops unexpectedly before the specified duration, check for errors in microphone access or WebSocket connection.