I finished a degree in computational linguistics last year and if it had not been for the scope and detail of this superb book I would have been completely lost. The authors can write well too - an entertaining text book would you believe? A word of warning. There are some basic errors in the mathematical expositions - which are careless and should have been ironed out. There is fairly good errata online, but it does pay to be wary about just blankly copying some of the maths. Keeps you on your toes! Those problems aside, there are so many goodies in this book that it cannot fail to get 5 stars. If you are doing any language processing, or you are embroiled in a semester long course, buy this book and save yourself a lot of grief.
This is the basic text for my master's course in Speech and Language Processing. It is used in several classes, though as the only text in the natural language processing class, which is clearly its strength. We also use it in the speech processing class, but it does need supplemental material here. If you are looking for a introduction to speech and language processing, you can't find a better book, and with a reasonable price at that. Specifically, the book covers natural language processing, computational linguistics and speech recognition. There is also a chapter dealing with speech synthesis, and another on machine translation. As a reminder of the important of linguistics in this field, even though it largely transcends it, the book is organised into four topical sections with several chapters each: Words, Syntax, Semantics and Pragmatics. The Words section includes chapters which introduce regular expressions, finite-state transducers, (computational) phonology, text-to-speech synthesis, probabilistic models of pronunciation and spelling, n-grams, and finally hidden Markov models (HMMs) and speech recognition. The Syntax section introduces word classes and part-of-speech tagging, context-free grammars, parsing, features and unification. The Semantics section has chapters on meaning representation, semantic analysis, lexical semantics, word sense disambiguation and information retrieval. Finally, the Pragmatics section covers discourse, dialogue, conversational agents, natural language generation and machine translation.
Doing a thesis on NLP, and checked this books out of the university library. Everything that one could want to know about linguistics, NLP, machine translation, it is all in here. Thouroughly recommended.