Hi everybody,
I have developed and tested a local speech recognition application on my Arduino VENTUNO Q using Arduino App Lab.
The application combines three Arduino components:
- Microphone API for audio recording.
- ASR brick for local speech-to-text transcription.
- WebUI brick for controlling the application from a browser.
The WebUI allows users to start and stop recording, display the recognized text, and copy it to the clipboard.
I tested the application in French, and automatic punctuation works well.
Some results from my tests:
- 18-second recording: 5.44 seconds for transcription.
- 13-second recording: 5.52 seconds for transcription.
The application runs locally on the VENTUNO Q.
The complete project, source code, documentation, and screenshot are available on GitHub:
https://github.com/philippe86220/VENTUNO-Q-Local-Speech-Recognition-with-Arduino-App-Lab
Many thanks to the Arduino team for providing these APIs and bricks.
I am very happy with what can already be achieved using Arduino App Lab on the VENTUNO Q!
Best regards.