The project is for a company called Neurolex Labs. Neurolex analyzes voice samples to help diagnose diseases such as schizophrenia, Parkinson’s disease, Alzheimer’s disease, and depression. In this case, we are analyzing human voice samples to determine the age of the speaker. We may want to know the age range of the individual so we can get specialized care for people in that age range. Moreover, knowing the age of the individual can give us a better context for the nature of their disease, and how likely they are to have a particular disease.
The name of the dataset is called the Common Voice Dataset by Mozilla. It contains about 200,000 voice samples. It contains the feature we are interested in, which is the age of the speaker. In order to featurize this audio data, we will be using two Python libraries: pyAudioAnalysis and Librosa.
For results and more detailed information on methodology, check the "Reports" folder