Classification of infant cry vocalizations using artificial neural networks (ANNs)

Petroni, M.; Malowany, A.S.; Johnston, Céleste; Stevens, Bonnie

doi:10.1109/icassp.1995.479734

Cited by 32 publications

(10 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All researchers who are working in this field have their own database. In [2,3], researchers have collected data from 16 healthy infants for cry classification. In [4] , cry is analyzed for hearing disorder detection with 37 infants out of which 14 had severe hearing loss and 23 had normal hearing.…”

Section: Introductionmentioning

confidence: 99%

Data collection and corpus design for analysis of nonnal and pathological infant cry

Chittora

Patil

2013

2013 International Conference Oriental COCOSDA Held Jointly With 2013 Conference on Asian Spoken Language Research and Evaluati

View full text Add to dashboard Cite

Cry is the first response of an infant as a result of interaction with external world. Till an infant learns to speak, it is the only communicating tool with which he/she can express his/her needs and ask for attention. Infant cry is an important tool for analysis of an infants' health condition. To understand the requirements of an infant (such as feeding, boredom, and health-related issues) acoustic analysis of an infant cry is essential. In this paper, the development of infant cry corpus is presented. Infant cry corpus is designed for the analysis of different types of cry modes, cry melodies, and assessment of cry from clinical diagnosis perspective. In this paper, along with database description spectrographic analysis for results is also reported on the designed corpus to signify the importance of cry as a tool for clinical analysis of infant health.

show abstract

Section: Introductionmentioning

confidence: 99%

Data collection and corpus design for analysis of nonnal and pathological infant cry

Chittora

Patil

2013

2013 International Conference Oriental COCOSDA Held Jointly With 2013 Conference on Asian Spoken Language Research and Evaluati

View full text Add to dashboard Cite

show abstract

“…To recognize the type of cry, there are two main processes have to be carried out; feature extraction and pattern classification [1][2][3]4]. Raw infant cry signals contain significant information, noise and redundant information.…”

Section: Introductionmentioning

confidence: 99%

“…One of the classification techniques namely Artificial Neural Network (ANN) is able to classify and group various types of cry [1][2][3]. Time Delay Neural Network (TDNN) has been used to differentiate cries between normal, deaf and asphyxiated infants.…”

Section: Introductionmentioning

confidence: 99%

The effect of F-ratio in the classification of asphyxiated infant cries using multilayer perceptron Neural Network

Zabidi

Mansor

Khuan

et al. 2010

2010 IEEE EMBS Conference on Biomedical Engineering and Sciences (IECBES)

View full text Add to dashboard Cite

Artificial Neural Network has been widely applied for solving pattern recognition problems including infant cry classification for detecting infant health and physical status. Feature extraction is usually performed using Mel Frequency Cepstrum Coefficient (MFCC) analysis. If irrelevant features in the MFCC are not removed, the performance of the MLP will be degraded. The use of F-ratio is essential to select the significant features. This paper examines the effect of selecting features using F-ratio on the classification accuracy of the MLP. Results obtained from direct selection of coefficients and selection of coefficients via F-ratio, were compared. It is found that the contribution of F-ratio in the selection of input for the MLP has managed to produce high classification accuracy.

show abstract

“…Due to their rich and unique nature, many types of intelligent classifiers can be used to classify them. In [1], Artificial Neural Network (ANN) was used to distinguish infant emotions based on their cry signals. Works by [2] has reported good performance of Time-Delay-Neural Network (TDNN) in classifying pathological states such as normal, deaf and asphyxiated infants, with up to 86.06% classification accuracy.…”

Section: Introductionmentioning

confidence: 99%

“…The optimal values for these parameters are critical in order to achieve best classifier accuracy. Works on audio feature extraction, generally use 20 to 40 filter banks, while the number of coefficients was in the range of 10 to 20 [1,2,[4][5][6][7][8][9][10][11]. However, there has been no systematic investigation on these selections, and how they affect classifier performance This paper describes the use of MLP to discriminate between healthy infants and infant suffering from hypothyroidism based on MFCC features extracted from the cry signals.…”

Section: Introductionmentioning

confidence: 99%

Investigation of Mel Frequency Cepstrum Coefficients parameters for classification of infant cries with hypothyroidism using MLP classifier

Zabidi

Mansor

Khuan

et al. 2010

The 2010 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Hypothyroidism in infants is caused by insufficient production of hormones by the thyroid gland. Due to stress in the chest cavity due to the enlarged liver, the cry signals are unique and can be distinguished from healthy infant cries. We investigate the usage of the Multilayer Perceptron (MLP) classifier to diagnose infant hypothyroidism. The Mel Frequency Cepstrum Coefficients (MFCC) feature extraction method was used to extract important information from the cry signal itself. This study investigates the number of filter banks and coefficients in MFCC to extract optimal information from infant cry signals, to be classified using MLP. The cry signals were first divided into equal-length segments, and MFCC was used to extract features from them. Tests on the combined University of Milano-Bicocca and Instituto Nacional de Astrofisica datasets yielded MLP classification accuracy of 89.18%, suggesting that the optimal MFCC resolution was obtained using 36 filter banks, and 19 coefficients.

show abstract

Classification of infant cry vocalizations using artificial neural networks (ANNs)

Cited by 32 publications

References 7 publications

Data collection and corpus design for analysis of nonnal and pathological infant cry

Data collection and corpus design for analysis of nonnal and pathological infant cry

The effect of F-ratio in the classification of asphyxiated infant cries using multilayer perceptron Neural Network

Investigation of Mel Frequency Cepstrum Coefficients parameters for classification of infant cries with hypothyroidism using MLP classifier

Contact Info

Product

Resources

About