Body Mass Index (BMI) Of Normal and Overweight/Obese Individuals Based on Speech Signals

Conventional method for measuring Body Mass Index (BMI) for individuals are using calibrated weight scale and measuring tape. However, there are certain cases in which the conventional method of measuring the BMI is not accessible. Thus, this experiment was proposed to overcome the problem using speech approach. In order to develop an effective BMI measuring system, speech signals of 30 subjects were recorded using a microphone. The dimension of the speech signal was reduced by extracting the relevant features using LPCC, MFCC and WPT based energy and entropy features. Lastly, both kNearest Neighbour (kNN) and Probabilistic Neural Network (PNN) were used to measure the BMI of an individual. The kNN classifier (97.50%) gives promising accuracy compared to PNN classifier (96.33%).

[1]  Jennifer R Harris,et al.  Sex differences in heritability of BMI: a comparative study of results from twin studies in eight countries. , 2003, Twin research : the official journal of the International Society for Twin Studies.

[2]  L. DeBruine,et al.  Vocal indicators of body size in men and women: a meta-analysis , 2014, Animal Behaviour.

[3]  Kuninori Shiwaku,et al.  The New BMI Criteria for Asians by the Regional Office for the Western Pacific Region of WHO are Suitable for Screening of Overweight to Prevent Metabolic Syndrome in Elder Japanese Workers , 2003, Journal of occupational health.

[4]  Yaacob Sazali,et al.  Comparison of LPCC and MFCC for isolated Malay speech recognition , 2012 .

[5]  Boncho Ku,et al.  A Novel Method for Classifying Body Mass Index on the Basis of Speech Signals for Future Clinical Applications: A Pilot Study , 2013, Evidence-based complementary and alternative medicine : eCAM.

[6]  Todor Ganchev,et al.  Estimation of unknown speaker’s height from speech , 2009, Int. J. Speech Technol..

[7]  Julio González,et al.  Research in acoustics of human speech sounds : Correlates and perception of speaker body size , 2007 .

[8]  M. Hariharan,et al.  A review: Malay speech recognition and audio visual speech recognition , 2012, 2012 International Conference on Biomedical Engineering (ICoBE).

[9]  R N Pierson,et al.  How useful is body mass index for comparison of body fatness across age, sex, and ethnic groups? , 1996, American journal of epidemiology.

[10]  Sazali Yaacob,et al.  Classification of speech dysfluencies with MFCC and LPCC features , 2012, Expert Syst. Appl..

[11]  Walker H. Land,et al.  A Multi-class Probabilistic Neural Network for Pathogen Classification , 2013, Complex Adaptive Systems.

[12]  M. Kalamani,et al.  Automatic Speech Recognition using ELM and KNN Classifiers , 2015 .

[13]  Philip Greenland,et al.  BMI and health-related quality of life in adults 65 years and older. , 2004, Obesity research.

[14]  M. Hariharan,et al.  Malay speech recognition in normal and noise condition , 2012, 2012 IEEE 8th International Colloquium on Signal Processing and its Applications.

[15]  Keun Ho Kim,et al.  Prediction of body mass index status from voice signals based on machine learning for automated medical applications , 2013, Artif. Intell. Medicine.

[16]  Donald F. Specht,et al.  Probabilistic neural networks , 1990, Neural Networks.

[17]  Keun Ho Kim,et al.  Study of a Vocal Feature Selection Method and Vocal Properties for Discriminating Four Constitution Types , 2012, Evidence-based complementary and alternative medicine : eCAM.

[18]  Sazali Yaacob,et al.  Objective evaluation of speech dysfluencies using wavelet packet transform with sample entropy , 2013, Digit. Signal Process..

[19]  Sazali Yaacob,et al.  Estimation of BMI status via speech signals using short-term cepstral features , 2015, 2015 IEEE International Conference on Control System, Computing and Engineering (ICCSCE).