This book takes an empirical approach to language processing, based on applying statistical and other machine-learning algorithms to large corporations.* Each chapter is built around one or more worked examples demonstrating the main idea of the chapter. * Uses worked examples to illustrate the relative strengths and weaknesses of various approaches. * Methodology boxes - Included in each chapter. * Introduces important methodological tools such as evaluation, wizard of oz techniques, etc. * Problem sets - Included in each chapter. * Integration of speech and text processing - Merges speech processing and natural language processing fields. * Empiricist/statistical/machine learning approaches to language processing - Covers all of the new statistical approaches, while still completely covering the earlier more structured and rule-based methods. * Includes modern rigorous evaluation metrics. * Unified and comprehensive coverage of the field - Covers the fundamental algorithms of various fields, whether originally proposed for spoken or written language. * Shows students how the same algorithm can be used for speech recognition and word-sense disambiguation. * Emphasis on Web and oth