Hybrid Neural Network Design and Implementation on FPGA for Infant Cry Recognition

Genetic Fuzzy Relational Neural Network for Infant Cry Classification

“…All experiments are binary classification. Although the problem of infant cry classification is indeed a multiclass problem, and our team has treated it in that way in several previous works( [8,9,10]), but for the present case we present a binary classification because our purpose is to compare our results with a particular similar work wich precissely had that binary approach. Results are compared with the work of Barajas and Reyes [2], that used the same databases.…”

“…For example, in the work of Suaste-Rivas et al [10] normal, deaf and asphyxia cries are classified with a FRNN obtaining results up to 88.00%. A later implementation of the FRNN on FPGA was done by Suaste-Rivas et al [8], also classifying normal, deaf and asphyxia cries and obtaining a generalization rate of 94.61%. It is important to point out that these works did not use n-fold cross validation; just randomly selected training and testing sets were used.…”

“…In [8,9,10] the FRNN was implemented for infant cry classification. Also, in [2] a proposal is presented to optimize the parameters for the FRNN using a genetic algorithm.…”

Classifying infant cry patterns by the Genetic Selection of a Fuzzy Model

“…Among these techniques, artificial neural networks have been one of the most widely used [13,26,41,45]. With the same purpose, the use of support vector machines (SVMs) [7,48], hidden Markov models [33,34], as well as several hybrid approaches that combine fuzzy logic with neural networks [44,[50][51][52], fuzzy logic with support vector machines [6] or evolutionary strategies with neural networks [23] have also been explored. Table 1 summarizes the characteristics of previous studies around the infant cry recognition.…”

Infant Cry Classification Using Genetic Selection of a Fuzzy Model

“…Among these techniques, artificial neural networks have been one of the most widely used [6,9,14,16]. It has also been explored the use of support vector machines [3,19], hidden Markov models [12,13], as well as several hybrid approaches that combine fuzzy logic with neural networks [15,20,21,22], fuzzy logic with support vector machines [2] or evolutionary strategies with neural networks [8]. These works have reported promising results in infant cry recognition.…”