Artificial neural networks in speech recogonization

Speech is the most effective form of interpersonal communication, which makes it an excellent choice for machine interfaces.This potential is shown by the growing need for automatic speech recognition systems.In this field, several approaches have been investigated, such as the Hidden Markov Model (HMM), Dynamic Time Warping (DTW), and Vector Quantization (VQ).However, the extensive use of artificial neural networks (ANN) in voice recognition is explored in this research.It carefully looks at the variety of neural network-based methods that can be used for speech recognition and compares their benefits and drawbacks.In the end, the study offers a recommendation for the best approach to further this field.