Development of Hausa Acoustic Model for Speech Recognition

Abstract

Acoustic modeling is essential for enhancing the accuracy of voice recognition software. To build an automatic speech system and application for any language, building an acoustic model is essential. In this regard, this research is concerned with the development of the Hausa acoustic model for automatic speech recognition. The goal of this work is to design and develop an acoustic model for the Hausa language. This is done by creating a word-level phonemes dataset from the Hausa speech corpus database. Then implement a deep learning algorithm for acoustic modeling. The model was built using Convolutional Neural Network that achieved 83% accuracy. The developed model can be used as a foundation for the development and testing of the Hausa speech recognition system.