Identifying familiar talkers over a 2.4 kbps LPC voice system

Abstract

A commonly cited drawback of narrow‐band digital voice communication systems such as the DoD standard LPC algorithm is that talker recognition is poor. Yet is is the opinion of many users that they frequently recognize the talker. Tape recordings of 24 talkers conversing over an unprocessed channel and over an LPC voice processing system were subjected to listening tests. The listeners were 24 co‐workers who listened to the tapes and attempted to identify each talker from a group of about 40 people working in the same branch. Prior to the recognition test, each of the listeners also rated his or her familiarity with each of the talkers and the distinctiveness of each talker's voice. There was some loss in voice recognition over LPC, but the recognition rate was still quite high. Unprocessed voices were correctly identified 88% of the time, whereas the same people talking over the LPC system were correctly identified 69% of the time. Talker familiarity was significantly correlated with correct identifications. There was no significant correlation between the rated distinctiveness of the talker and correct identifications. However, familiarity and distinctiveness ratings were highly correlated. This suggests that people consider a familiar voice to be distinctive regardless of other characteristics that might make that particular voice stand out in a crowd. [Work supported by NAVELEX.]