Yorùbá Gender Recognition from Speech Using Neural Networks

Abstract

The impressive improvement in performance obtained using neural networks for automatic speech recognition (ASR) have motivated the application of neural networks to other speech technologies such as speaker, emotion, language, and gender recognition. Prior work has shown significant improvement in gender recognition from images and videos. This paper uses speech to build a gender recognition system based on neural networks. Three types of neural networks are investigated to find the best model for gender recognition system using Yoruba, namely, feed-forward artificial neural networks (Multilayer Perceptrons), Recurrent neural networks (long short-term memory), and Convolutional neural networks. All the classifier models obtained the state-of-the-art performance in speech-based gender recognition with 99% in accuracy and F 1 score.