This is a simple web app that uses an algorithm to detect vocal fry in real-time audio. Click the "Start" button to begin detecting vocal fry. The screen will flash red when vocal fry is detected.
Currently the model was only trained on one (my own) voice, so it may not work well for other voices. I plan to train the model on a larger dataset in the future.
There are also some issues with one off false positives, and the model is not perfect. It is a work in progress.
Lastly, I have not tested this in noisy environments, and I suspect there will be even more false positives in that case.