Speech-language pathologists traditionally assess customers’ speaking ability before prescribing treatment, which will then require certain pieces of equipment used for therapy. The assessment could be costly and inconvenient for accessing communities in the remotest locations. In contrast, AVDs represent more cost-efficient mechanisms that achieve repeatability, are non-invasive, and further reduce frequent clinical visits. In terms of study, three popular acoustic feature extraction techniques, such as PLP, LPCC, and MFCC, are applied in classifying voice pathologies. A comparative analysis between the hybrid quantum classifier and the one-dimensional convolutional neural network (1D-CNN) is proposed for classification. The experiments are executed in the Saarbrücken Voice Disorders (SVD) database. The results show that the CNN classifier performs better than the hybrid quantum approach for classification when considering 1D feature representations.