Learning Disability (LD) is an umbrella term for various learning difficulties. It is a neurological disorder that affects the brain of children and hinders their capacity to do one or more specified activities. Youngsters with learning disabilities are neither sluggish nor mentally retarded. This paper utilizes Speech, as an acceptable motor activity to address the problem and its solution. In this work, Mel-Frequency Cepstral Coefficients (MFCC) and Linear Predictive Cepstral Coefficients (LPCC) are used to extract speech characteristics. For every speech sample, 205 characteristics are retrieved, consisting of 13 acoustical features from MFCC, 128 features from LPCC, and 64 features from power spectrum density. In addition, linear predictive coding-based parameterization is used to every speech sample to extract Weighted Linear Predictive Cepstral Coefficients (WLPCC) and Linear Predictive Coding features (LPC). The Extracted Features are identified using a Deep Convolutional Neural Network (DCNN), and their performance is analyzed. It is found that the correlation between speech characteristics and various classification models may be assessed to get the highest classification accuracy. These findings demonstrate that the suggested feature extraction approaches and classification models are highly successful in categorizing the speech of intellectually impaired and typically developing youngsters.