On investigating efficient methodology for environmental sound recognition

Cruz Alfredo Ruiz-Martinez, Muhammad Tahir Akhtar, Yoshikazu Washizawa, Enrique Escamilla-Hernandez

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

This paper presents a comparative study of various methods to identify the environmental sounds. We evaluate two methods for feature extraction: Mel Frequency Cepstral Coefficients (MFCC) which is well known for speaker identification, and Matching Pursuit (MP) with Gabor Dictionary which gives a time frequency representation employed for scene recognition. In the classification stage, we show a comparison among Support Vector Machines (SVM), Logistic Regression, and Backpropagation Artificial Neural Network (BP-ANN). Simulation results show that MFCC gives a higher recognition performance as compared with MP. Furthermore, by concatenating MFCC features with some feature of MP, e.g., scale, might also improve performance in some situations. We observe that SVM show the best performance among the classifiers, for clean as well noisy signals.

Original languageEnglish
Title of host publicationISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems
Pages210-214
Number of pages5
DOIs
Publication statusPublished - Dec 1 2013
Event2013 21st International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2013 - Naha, Okinawa, Japan
Duration: Nov 12 2013Nov 15 2013

Publication series

NameISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems

Conference

Conference2013 21st International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2013
CountryJapan
CityNaha, Okinawa
Period11/12/1311/15/13

Fingerprint

Support vector machines
Acoustic waves
Glossaries
Backpropagation
Logistics
Feature extraction
Classifiers
Neural networks

ASJC Scopus subject areas

  • Artificial Intelligence
  • Signal Processing

Cite this

Ruiz-Martinez, C. A., Akhtar, M. T., Washizawa, Y., & Escamilla-Hernandez, E. (2013). On investigating efficient methodology for environmental sound recognition. In ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems (pp. 210-214). [6704548] (ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems). https://doi.org/10.1109/ISPACS.2013.6704548

On investigating efficient methodology for environmental sound recognition. / Ruiz-Martinez, Cruz Alfredo; Akhtar, Muhammad Tahir; Washizawa, Yoshikazu; Escamilla-Hernandez, Enrique.

ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems. 2013. p. 210-214 6704548 (ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ruiz-Martinez, CA, Akhtar, MT, Washizawa, Y & Escamilla-Hernandez, E 2013, On investigating efficient methodology for environmental sound recognition. in ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems., 6704548, ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems, pp. 210-214, 2013 21st International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2013, Naha, Okinawa, Japan, 11/12/13. https://doi.org/10.1109/ISPACS.2013.6704548
Ruiz-Martinez CA, Akhtar MT, Washizawa Y, Escamilla-Hernandez E. On investigating efficient methodology for environmental sound recognition. In ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems. 2013. p. 210-214. 6704548. (ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems). https://doi.org/10.1109/ISPACS.2013.6704548
Ruiz-Martinez, Cruz Alfredo ; Akhtar, Muhammad Tahir ; Washizawa, Yoshikazu ; Escamilla-Hernandez, Enrique. / On investigating efficient methodology for environmental sound recognition. ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems. 2013. pp. 210-214 (ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems).
@inproceedings{6ba9f567aff445559dc3bb53220d4bbf,
title = "On investigating efficient methodology for environmental sound recognition",
abstract = "This paper presents a comparative study of various methods to identify the environmental sounds. We evaluate two methods for feature extraction: Mel Frequency Cepstral Coefficients (MFCC) which is well known for speaker identification, and Matching Pursuit (MP) with Gabor Dictionary which gives a time frequency representation employed for scene recognition. In the classification stage, we show a comparison among Support Vector Machines (SVM), Logistic Regression, and Backpropagation Artificial Neural Network (BP-ANN). Simulation results show that MFCC gives a higher recognition performance as compared with MP. Furthermore, by concatenating MFCC features with some feature of MP, e.g., scale, might also improve performance in some situations. We observe that SVM show the best performance among the classifiers, for clean as well noisy signals.",
author = "Ruiz-Martinez, {Cruz Alfredo} and Akhtar, {Muhammad Tahir} and Yoshikazu Washizawa and Enrique Escamilla-Hernandez",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/ISPACS.2013.6704548",
language = "English",
isbn = "9781467363617",
series = "ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems",
pages = "210--214",
booktitle = "ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems",

}

TY - GEN

T1 - On investigating efficient methodology for environmental sound recognition

AU - Ruiz-Martinez, Cruz Alfredo

AU - Akhtar, Muhammad Tahir

AU - Washizawa, Yoshikazu

AU - Escamilla-Hernandez, Enrique

PY - 2013/12/1

Y1 - 2013/12/1

N2 - This paper presents a comparative study of various methods to identify the environmental sounds. We evaluate two methods for feature extraction: Mel Frequency Cepstral Coefficients (MFCC) which is well known for speaker identification, and Matching Pursuit (MP) with Gabor Dictionary which gives a time frequency representation employed for scene recognition. In the classification stage, we show a comparison among Support Vector Machines (SVM), Logistic Regression, and Backpropagation Artificial Neural Network (BP-ANN). Simulation results show that MFCC gives a higher recognition performance as compared with MP. Furthermore, by concatenating MFCC features with some feature of MP, e.g., scale, might also improve performance in some situations. We observe that SVM show the best performance among the classifiers, for clean as well noisy signals.

AB - This paper presents a comparative study of various methods to identify the environmental sounds. We evaluate two methods for feature extraction: Mel Frequency Cepstral Coefficients (MFCC) which is well known for speaker identification, and Matching Pursuit (MP) with Gabor Dictionary which gives a time frequency representation employed for scene recognition. In the classification stage, we show a comparison among Support Vector Machines (SVM), Logistic Regression, and Backpropagation Artificial Neural Network (BP-ANN). Simulation results show that MFCC gives a higher recognition performance as compared with MP. Furthermore, by concatenating MFCC features with some feature of MP, e.g., scale, might also improve performance in some situations. We observe that SVM show the best performance among the classifiers, for clean as well noisy signals.

UR - http://www.scopus.com/inward/record.url?scp=84894111002&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84894111002&partnerID=8YFLogxK

U2 - 10.1109/ISPACS.2013.6704548

DO - 10.1109/ISPACS.2013.6704548

M3 - Conference contribution

SN - 9781467363617

T3 - ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems

SP - 210

EP - 214

BT - ISPACS 2013 - 2013 International Symposium on Intelligent Signal Processing and Communication Systems

ER -