Ebook: Robust Speech Recognition and Understanding

InTech, 2007. — 470 p.Digital speech processing is a major field in current research all over the world. In particular for automatic speech recognition (ASR). Very significant achievements have been made since the first attempts of digit recognizers in the 1950’s and 1960’s when spectral resonances were determined by analogue filters and logical circuits. As prof. Furui pointed out in his review on 50 years of automatic speech recognition at the 32nd IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2007, we may now see speech recognition systems in their 3.5th generation. Although there are many excellent systems for continuous speech recognition, speech translation and information extraction, ASR systems need to be improved for spontaneous speech. Furthermore, robustness under noisy conditions is still a goal that has not been achieved entirely if distant microphones are used for speech input. The automated recognition of emotion is another aspect in voice-driven systems that has gained much importance in recent years. For natural language understanding in general, and for the correct interpretation of a speaker’s recognized words, such paralinguistic information may be used to improve future speech systems.
This book on Robust Speech Recognition and Understanding brings together many different aspects of the current research on automatic speech recognition and language understanding. The first four chapters address the task of voice activity detection which is considered an important issue for all speech recognition systems. The next chapters give several extensions to state-of-the-art HMM methods. Furthermore, a number of chapters particularly address the task of robust ASR under noisy conditions. Two chapters on the automatic recognition of a speaker’s emotional state highlight the importance of natural speech understanding and interpretation in voice-driven systems. The last chapters of the book address the application of conversational systems on robots, as well as the autonomous acquisition of vocalization skills.
We want to express our thanks to all authors who have contributed to this book by the best of their scientific work. We hope you enjoy reading this book and get many helpful ideas for your own research or application of speech technology.