The following demo introduces gender and voice sentiment recognition with deep learning.

Voice classification model: CNN + CNN
Classes: female / male
Classes: angry / disgusted / frightened / happy / neutral / sad / surprised
Source: Github

Steps:

1. The demo can be started by accepting the Adatkezelési Tájékoztató/Privacy Policy below. 
After click on button “OK!“, your browser will ask for permission to use your microphone.

2. Press button “Record” which will record a maximum 4 seconds long speech of yours.
Make sure you use simple sentences.

3. Press button “Predict“. then AI will try to figure out your gender(male/female) and voice sentiment: angry, disgusted, frightened, happy,  neutral, sad or surprised.