Expert methods of foreign-language speakers’ identification by speech prosody

Бесплатный доступ

According to the data received from the surveys of the criminal environment in Russia over recent years, there is a substantial growth of the crimes committed by the Tadjik, Uzbek and criminal groups of other ethnical origin. In forensic phonetics, expert witnesses might be confronted with voice samples for auditory evaluation from a language they do not understand. In speaker identification experiments, it has been shown that knowledge of the target language affects recognition results. The present paper deals with possibility of primary recognition of people speaking foreign language or foreign-accented Russian for the purpose of their further voice identification. This paper also presents one of the automatically-assisted methods of speaker identification i.e. syntagma pitch pattern analysis method (on the ground of pitch pattern segments comparison). There is a brief description of the methodological background and main components of speech melodic structure. The author compares prosodic organization of utterances in speech behaviour of Russian speakers and speakers of foreign languages, and defines the properties of adequate understanding of the meaning of the utterance and of the speaker’s communicative goal, including situations when the listener is not a native speaker of the language being used. The results of this study show that (a) experts with experience obtain significantly better recognition results than experts without work experience, (b) no clear evidence was found that the typological difference between the native language of the listener and the target language influenced recognition performance.

Еще

Tajik language, the gipsy language, speech interference, foreign speech perception, voice and speech identification, syntagma, melodic patterns of syntagmas, prosodic parameters of speech

Короткий адрес: https://sciup.org/149129914

IDR: 149129914   |   DOI: 10.15688/jvolsu2.2018.3.6

Статья научная