Angielski
This book covers the state-of-the-art in noise robustness for deep neural-network-based speech recognition with a focus on applications in distant speech. It provides insights and detailed descriptions of some of the key technologies in the field, including speech enhancement, neural-network-based noise reduction, robust features, acoustic model adaptation, training data augmentation, novel network topologies, and training criteria. The contributed chapters also include descriptions of benchmark tools and datasets widely used in the field. This book is intended for researchers and practitioners working in the field of automatic speech recognition who are engaged in improving noise robustness. It will also be of interest to graduate students in electrical engineering or computer science, who will find it a useful guide to this field of research.
