Applying feature extraction of speech recognition on VoIP auditing

This paper presents an approach that combines both VoIP auditing and speech recognition, and an application of using speech recognition on VoIP voice data is described. It also deals with how to obtain VoIP voice data during audio communication: signaling protocol analysis and voice data traffic analysis. Feature extraction is a very important step in speech recognition, and the result derived from this research shows that the differences between the features extracted from voice recorded directly and voice captured through VoIP are very small and that applying speech recognition on VoIP auditing is applicable.