Machine Learning in Automatic Speech Recognition: A Survey

Over the past few decades, there has been tremendous development in machine learning paradigms used in automatic speech recognition (ASR) for home automation to space exploration. Though commercial speech recognizers are available for certain well-defined applications like dictation and transcription, many issues in ASR like recognition in noisy environments, multilingual recognition, and multi-modal recognition are yet to be addressed effectively. A comprehensive review of common machine learning techniques like artificial neural networks, support vector machines, and Gaussian mixture models along with hidden Markov models employed in ASR is provided. A thorough review on the recent developments in deep learning which has provided significant improvements in ASR performance, along with its relevance in the future of ASR, is also presented.