Intoxication Detection using Audio

Intoxication detection using speech involves analyzing an individual’s speech patterns to determine if they are intoxicated. Such trained models can reduce the usage of unhygienic devices like breathalyzers. This is generally done by examining numerous aspects of utterance and acoustics of speech signals used for training MFCC and CNN models. The primary objective is to develop a non- invasive, accurate approach to intoxication detection that can be deployed for law enforcement and occupational intoxication testing.