Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Recolector de Cienci...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
ResearchGate Data
Preprint . 2019
Data sources: Datacite
versions View all 2 versions
addClaim

Sistema multiparamétrico para la comparación forense de hablantes

Authors: San Segundo Fernández, Eugenia; Univaso, Pedro; Gurlekian, Jorge;

Sistema multiparamétrico para la comparación forense de hablantes

Abstract

[ES] En la comparación forense de hablantes se pueden examinar diversos parámetros. En el presente trabajo se utilizó un sistema multiparamétrico que combina parámetros acústicos de largo plazo (F0, cualidad de voz y aspectos duracionales) con los parámetros de corto plazo (MFCC) empleados por un sistema automático estándar basado en el enfoque i-vector/PLDA (sistema base). El objetivo era determinar si el nuevo sistema de comparación de hablantes ofrece mejor rendimiento que el sistema base. Para ello se llevaron a cabo tres experimentos condiseños diferentes –que permitieron evaluar el nuevo sistema multiparamétrico en condiciones extremas, a modo de prueba de estrés–: (1) uso de grabaciones con características forenses realistas (p. ej. ruido de fondo, reverberación, variabilidad intra-hablante, compresión de la señal); (2) comparación de las voces de 12 parejas de gemelos monocigóticos; y (3) cotejo de voces con enmascaramiento mediante pinzamiento de nariz. Los resultados obtenidos con el nuevo sistema muestran una mejora de rendimiento con respecto al sistema base, si bien el aporte medio de los parámetros de largo plazo al nuevo sistema fue de un 6.5%, siendo el restante 93.5% responsabilidad de los parámetros de corto plazo.

[EN] In Forensic Speaker Comparison (FSC) several different parameters are commonly analysed. In this investigation we propose a multiparametric system combining long-term features (f0, voice quality and durational aspects) with short-term features (MFCCs), used by a standard automatic system based on i-vector/PLDA approaches (baseline system). The objective was to determine if the performance of the new FSC system is better than that of the baseline system. For this, three experimental designs were carried out –allowing us to evaluate the new multiparametric system in extreme conditions, as if it was a stress test–: (1) use of forensically-realistic characteristics (e.g. background noise, reverberation, intra-speaker variability, signal compression); (2) voice comparison of 12 monozygotic twin pairs; and (3) comparison of disguised voices through noise pinching. The results show that the new system performs better than the baseline system although the mean contribution of long-term features to the new system was 6.5%, with the short-term features being responsible for the remaining 93.5%.

Este artículo está sujeto a una licencia CC BY NC ND 4.0

Peer reviewed

Keywords

Stress test, Gemelos, Disguise, Twins, Linguistic research, Disimulo, Voice quality, Prueba de estrés, MFCCs, Calidad de voz

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
    OpenAIRE UsageCounts
    Usage byUsageCounts
    visibility views 61
    download downloads 40
  • 61
    views
    40
    downloads
    Powered byOpenAIRE UsageCounts
Powered by OpenAIRE graph
Found an issue? Give us feedback
visibility
download
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
views
OpenAIRE UsageCountsViews provided by UsageCounts
downloads
OpenAIRE UsageCountsDownloads provided by UsageCounts
0
Average
Average
Average
61
40
Green