Название: Extraction of Prosody for Automatic Speaker, Language, Emotion and Speech Recognition, 2nd Edition Автор: Leena Mary
Издательство: Springer
Серия: SpringerBriefs in Speech Technology. Studies in Speech Signal Processing, Natural Language Understanding, and Machine Learning
Год: 2018 (2019 edition)
Язык: английский
Формат: pdf (true), epub
Размер: 10.17 MB
This updated book expands upon prosody for recognition applications of speech processing. It includes importance of prosody for speech processing applications; builds on why prosody needs to be incorporated in speech processing applications; and presents methods for extraction and representation of prosody for applications such as speaker recognition, language recognition and speech recognition. The updated book also includes information on the significance of prosody for emotion recognition and various prosody-based approaches for automatic emotion recognition from speech.
Human beings recognize speaker, language, emotion, and speech using multiple cues present in speech signal and evidences are combined to arrive at a decision. Humans use several prosodic cues for these recognition tasks. But conventional automatic speaker, language, emotion, and speech recognition systems mostly rely on spectral/cepstral features which are affected by channel mismatch and noise. Therefore incorporation of prosody into these automatic recognition tasks will make them more robust and human like. In recent years there has been an increasing interest in using prosody for various speech processing applications. This book focuses on extraction and representation of prosodic features directly from speech signal for speaker, language, emotion, and speech recognition. It is organized into three chapters. The first chapter describes the significance of prosody for speaker recognition, language recognition, emotion recognition, and speech recognition. The second chapter explains various methods for the automatic extraction and representation of prosody for the above applications. The third chapter deals with modeling of prosody and describes methods used for the integration of prosodic knowledge into conventional recognition systems. The discussions are limited to selected methods that use direct extraction of prosody from speech signal, eliminating the need for hand annotation of prosodic events.
The material presented in this book is primarily intended for the speech processing researchers and for those who develop software for speech processing applications.
Скачать Extraction of Prosody for Automatic Speaker, Language, Emotion and Speech Recognition, 2nd Edition