Phonological Parsing in Speech Recognition describes a novel recognizer that is designed to exploit variant cues by parsing the input utterance into syllables and other suprasegmental constituents using phrase-structure parsing techniques. The unique matrix view of parsing on engineering grounds allows for a number of constructions to be defined in natural ways. The interdisciplinary approach presented by the author makes this book a valuable reference to both researchers and students in speech recognition and linguistics.

Скачать книгу бесплатно (djvu, 10.05 Mb)

---

Читать «Phonological parsing in speech recognition»