Un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formanti

In this paper, a method for the forensic identification of speakers is presented, based on the analysis of the pitch and the first three formants of the four vowel: "a", "e", "i" and "o". Using these data, the method estimates the probability density function...

Full description

Bibliographic Details
Main Authors: Tommaso Bove, Paolo Emilio Giua, Alessandra Forte, Carla Rossi
Format: Article
Language:English
Published: University of Bologna 2007-10-01
Series:Statistica
Online Access:http://rivista-statistica.unibo.it/article/view/420
id doaj-ce5843e835094a23ac8dd3abdb31d236
record_format Article
spelling doaj-ce5843e835094a23ac8dd3abdb31d2362020-11-24T22:39:35ZengUniversity of BolognaStatistica0390-590X1973-22012007-10-0162347549010.6092/issn.1973-2201/420411Un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formantiTommaso BovePaolo Emilio GiuaAlessandra ForteCarla RossiIn this paper, a method for the forensic identification of speakers is presented, based on the analysis of the pitch and the first three formants of the four vowel: "a", "e", "i" and "o". Using these data, the method estimates the probability density function (pdf) of the Mahalanobis distance both of the defendant from himself (intra-distance estimation) and from the voices of the control set (inter-distance estimation), using the Kernel method for each vowel. The Mahalanobis norm is then used to estimate the pdf related to the four vowel. The sample under study is then classified according to the Maximum Likelihood Criterion approach. This allows one to estimate a unique decision threshold and the probabilities of the two possible classification errors (false acceptance and false rejection). The method has been applied to real data provided by Police Scientific Service of Rome, in the framework of a European Project.http://rivista-statistica.unibo.it/article/view/420
collection DOAJ
language English
format Article
sources DOAJ
author Tommaso Bove
Paolo Emilio Giua
Alessandra Forte
Carla Rossi
spellingShingle Tommaso Bove
Paolo Emilio Giua
Alessandra Forte
Carla Rossi
Un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formanti
Statistica
author_facet Tommaso Bove
Paolo Emilio Giua
Alessandra Forte
Carla Rossi
author_sort Tommaso Bove
title Un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formanti
title_short Un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formanti
title_full Un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formanti
title_fullStr Un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formanti
title_full_unstemmed Un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formanti
title_sort un metodo statistico per il riconoscimento del parlatore basato sull'analisi dei formanti
publisher University of Bologna
series Statistica
issn 0390-590X
1973-2201
publishDate 2007-10-01
description In this paper, a method for the forensic identification of speakers is presented, based on the analysis of the pitch and the first three formants of the four vowel: "a", "e", "i" and "o". Using these data, the method estimates the probability density function (pdf) of the Mahalanobis distance both of the defendant from himself (intra-distance estimation) and from the voices of the control set (inter-distance estimation), using the Kernel method for each vowel. The Mahalanobis norm is then used to estimate the pdf related to the four vowel. The sample under study is then classified according to the Maximum Likelihood Criterion approach. This allows one to estimate a unique decision threshold and the probabilities of the two possible classification errors (false acceptance and false rejection). The method has been applied to real data provided by Police Scientific Service of Rome, in the framework of a European Project.
url http://rivista-statistica.unibo.it/article/view/420
work_keys_str_mv AT tommasobove unmetodostatisticoperilriconoscimentodelparlatorebasatosullanalisideiformanti
AT paoloemiliogiua unmetodostatisticoperilriconoscimentodelparlatorebasatosullanalisideiformanti
AT alessandraforte unmetodostatisticoperilriconoscimentodelparlatorebasatosullanalisideiformanti
AT carlarossi unmetodostatisticoperilriconoscimentodelparlatorebasatosullanalisideiformanti
_version_ 1725708179173539840