This paper presents a study on the performance of a speaker identification system in challenging environmental conditions, such as in the presence of noise and at a distance. We propose a robust speaker recognition algorithm, which includes a smart pre-processing method based on Voice Activity Detection (VAD), capable to boost the system accuracy. Results show that our solution is able to improve the correct classification rate of traditional speaker recognition systems, even in case of distant audio acquisition and noisy environments.

Performance analysis of smart audio pre-processing for noise-robust text-independent speaker recognition

Bisio, Igor;Garibotto, Chiara;Sciarrone, Andrea;Lavagetto, Fabio
2018

Abstract

This paper presents a study on the performance of a speaker identification system in challenging environmental conditions, such as in the presence of noise and at a distance. We propose a robust speaker recognition algorithm, which includes a smart pre-processing method based on Voice Activity Detection (VAD), capable to boost the system accuracy. Results show that our solution is able to improve the correct classification rate of traditional speaker recognition systems, even in case of distant audio acquisition and noisy environments.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/923593
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
social impact