Normal view MARC view ISBD view

Linguistic and forensic phonetics identification of a male speaker (the real case) / Varošanec-Škarić, Gordana ; Kišiček, Gabrijela.

By: Varošanec-Škarić, Gordana.
Contributor(s): Kišiček, Gabrijela [aut].
Material type: ArticleArticleDescription: 34-34 str.Subject(s): 6.03 | speaker identification, forensic phonetics, auditory and acoustic analyisis eng In: 19th Annual Conference of IAFPA (18-21.07. 2010. ; Trier, Njemačka) Abstracts for the 19th Annual Conference of the International Association for Forensic Phonetics and Acoustics str. 34-34Braun, AngelikaSummary: Based on the demand of Croatian Court one male voice was to be identified on the recordings made during police surveillance from the GSM phone. Male voice had to be identified form the recording 1 - duration of 2, 16 min (CD 1) and compared with male voice on the recording 2 (CD 2) duration 7, 39 min. With the intent to discover elements of regional speech, dialect and idiolect, phonetic transcription was made for the whole speech and was acoustically analyzed. Linguistic analysis showed higher absolute and relative frequency of idiosyncratic words like jaran and bolan compared with the frequency of the same words in Croatian language. Identified speaker has Shtokavian, Ikavian, Bosnian origin (recognizable accents belong to Eastern variety, accent shifting, specific dialectical vowel pronunciation). Vowels are closed, more front and more closed than standard Croatian pronunciation, reduced and syncopated). Phonetic description of voice shows lightly elevated voice, close span of jaw, denasalized and modal type voice. Formant analysis of stressed vowels showed the difference (Z-test) between the identified speaker and standard Croatian pronunciation (for lower F1 /a/: p = 0, 001, F2higher for /a/, /e/, /o/, /u/). The results also showed the difference between the identified speaker and the speaker used for the comparison (significantly higher F2 for /i/, and lower F3 for /o/). Long term average spectra show timbre difference between two speakers.
Tags from this library: No tags from this library for this title. Log in to add tags.
No physical items for this record

Based on the demand of Croatian Court one male voice was to be identified on the recordings made during police surveillance from the GSM phone. Male voice had to be identified form the recording 1 - duration of 2, 16 min (CD 1) and compared with male voice on the recording 2 (CD 2) duration 7, 39 min. With the intent to discover elements of regional speech, dialect and idiolect, phonetic transcription was made for the whole speech and was acoustically analyzed. Linguistic analysis showed higher absolute and relative frequency of idiosyncratic words like jaran and bolan compared with the frequency of the same words in Croatian language. Identified speaker has Shtokavian, Ikavian, Bosnian origin (recognizable accents belong to Eastern variety, accent shifting, specific dialectical vowel pronunciation). Vowels are closed, more front and more closed than standard Croatian pronunciation, reduced and syncopated). Phonetic description of voice shows lightly elevated voice, close span of jaw, denasalized and modal type voice. Formant analysis of stressed vowels showed the difference (Z-test) between the identified speaker and standard Croatian pronunciation (for lower F1 /a/: p = 0, 001, F2higher for /a/, /e/, /o/, /u/). The results also showed the difference between the identified speaker and the speaker used for the comparison (significantly higher F2 for /i/, and lower F3 for /o/). Long term average spectra show timbre difference between two speakers.

Projekt MZOS 130-0000000-0786

ENG

There are no comments for this item.

Log in to your account to post a comment.

Powered by Koha

//