Abstract A dialogue algorithm is described that executes Prolog-style rules in an attempt to achieve a goal. The algorithm selects paths in the proof in an attempt to achieve success and proves subgoals on the[…]
Abstract Acquiring a language depends on correctly recovering words from fluent speech. Thus every language learner needs to solve the segmentation problem. Infants appear to have some success in segmenting speech even during their first[…]
Abstract In this talk, I will present a model of text analysis for text-to-speech synthesis (TTS) based on (weighted) finite-state transducers — WFSTs, which serves as the text-analysis module of the multilingual Bell Labs TTS[…]
Abstract In most current speech recognizers, the observation space of an utterance consists of a temporal sequence of “frames”. An important property of this framework is that every segmentation of the input utterance accounts for[…]
Abstract Sociolinguistic research over the past two decades finds linguistic change in progress in the vowel systems of all the large cities of North America, leading to an increasing diversification of English dialects through mergers[…]
Abstract This talk introduces two recent topics in speech recognition research at Furui Research Laboratory, NTT Human Interface Laboratories.The first topic is large-vocabulary continuous speech recognition (LVCSR) using Japanese business newspaper task. No research efforts[…]
Abstract We explore the concept of designing parsers for the specific problem at hand. We consider two different classes of problems. First, we consider designing parsers which perform well on a given evaluation metric. While[…]
