Influence of Noise on the Effectiveness of Speaker Identification in the Acoustics of Crime
×
Cite This
Citation & Abstract
T. Smutnicki, and S. Brachmanski, "Influence of Noise on the Effectiveness of Speaker Identification in the Acoustics of Crime," Paper 9333, (2015 May.). doi:
T. Smutnicki, and S. Brachmanski, "Influence of Noise on the Effectiveness of Speaker Identification in the Acoustics of Crime," Paper 9333, (2015 May.). doi:
Abstract: One of the main elements of the research in acoustics of crime is to compare the evidential recording with the comparative adequate pattern. Unfortunately, the evidential recording usually has poor quality and contains relatively high level of noise, which results from the way of its acquiring, namely eavesdropping or record of automatic monitoring. The signal quality and the noise to signal ratio have an impact on the value of the extracted voice metrics. In this paper we analyze factors that may have an impact on formants value in the human voice. Based on Six Sigma methodology we also designed and performed an experiment that allowed us to determine the extent in which various factors influence on the resulting parameters.
@article{smutnicki2015influence,
author={smutnicki, tomasz and brachmanski, stefan},
journal={journal of the audio engineering society},
title={influence of noise on the effectiveness of speaker identification in the acoustics of crime},
year={2015},
volume={},
number={},
pages={},
doi={},
month={may},}
@article{smutnicki2015influence,
author={smutnicki, tomasz and brachmanski, stefan},
journal={journal of the audio engineering society},
title={influence of noise on the effectiveness of speaker identification in the acoustics of crime},
year={2015},
volume={},
number={},
pages={},
doi={},
month={may},
abstract={one of the main elements of the research in acoustics of crime is to compare the evidential recording with the comparative adequate pattern. unfortunately, the evidential recording usually has poor quality and contains relatively high level of noise, which results from the way of its acquiring, namely eavesdropping or record of automatic monitoring. the signal quality and the noise to signal ratio have an impact on the value of the extracted voice metrics. in this paper we analyze factors that may have an impact on formants value in the human voice. based on six sigma methodology we also designed and performed an experiment that allowed us to determine the extent in which various factors influence on the resulting parameters.},}
TY - paper
TI - Influence of Noise on the Effectiveness of Speaker Identification in the Acoustics of Crime
SP -
EP -
AU - Smutnicki, Tomasz
AU - Brachmanski, Stefan
PY - 2015
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - May 2015
TY - paper
TI - Influence of Noise on the Effectiveness of Speaker Identification in the Acoustics of Crime
SP -
EP -
AU - Smutnicki, Tomasz
AU - Brachmanski, Stefan
PY - 2015
JO - Journal of the Audio Engineering Society
IS -
VO -
VL -
Y1 - May 2015
AB - One of the main elements of the research in acoustics of crime is to compare the evidential recording with the comparative adequate pattern. Unfortunately, the evidential recording usually has poor quality and contains relatively high level of noise, which results from the way of its acquiring, namely eavesdropping or record of automatic monitoring. The signal quality and the noise to signal ratio have an impact on the value of the extracted voice metrics. In this paper we analyze factors that may have an impact on formants value in the human voice. Based on Six Sigma methodology we also designed and performed an experiment that allowed us to determine the extent in which various factors influence on the resulting parameters.
One of the main elements of the research in acoustics of crime is to compare the evidential recording with the comparative adequate pattern. Unfortunately, the evidential recording usually has poor quality and contains relatively high level of noise, which results from the way of its acquiring, namely eavesdropping or record of automatic monitoring. The signal quality and the noise to signal ratio have an impact on the value of the extracted voice metrics. In this paper we analyze factors that may have an impact on formants value in the human voice. Based on Six Sigma methodology we also designed and performed an experiment that allowed us to determine the extent in which various factors influence on the resulting parameters.
Authors:
Smutnicki, Tomasz; Brachmanski, Stefan
Affiliation:
Wroclaw University of Technology, Wroclaw, Poland
AES Convention:
138 (May 2015)
Paper Number:
9333
Publication Date:
May 6, 2015Import into BibTeX
Subject:
Applications in Audio
Permalink:
http://www.aes.org/e-lib/browse.cfm?elib=17757