Modeling disfluencies in conversational speech
نویسندگان
چکیده
Conversational speech is notably di erent from read speech in several ways, particularly in the presence of dis uencies but also in the frequent use of a small set of words that mark the ow of the discourse. Dis uencies are sometimes viewed as a \problem" in language modeling, where most previous work has focused on written text. In this paper, we take the view that dis uencies provide information themselves. In particular, we give evidence that lled pauses serve di erent functions, including marking linguistic unit and restart boundaries, and signaling hesitation where the speaker wants to hold the oor. The di erent functions can be connected to similar functions of other words common in spontaneous but not written speech, and the particular function a ects the word conditioning choices in a variable ngram model. Thus, at least some of the idiosyncrasies of spontaneous speech can be viewed as a source of information for language modeling rather than an interruption in the linguistic structure.
منابع مشابه
Automatic Detection of Sentence Boundaries, Disfluencies, and Conversational Fillers in Spontaneous Speech
Automatic Detection of Sentence Boundaries, Disfluencies, and Conversational Fillers in Spontaneous Speech
متن کاملModeling of Pronunciation, Language and Nonverbal Units at Conversational Russian Speech Recognition
The main problems of a conversational Russian speech recognition system development are variability of pronunciation, free word-order in sentences and presence of speech disfluencies. In the paper, pronunciation variability is modeled by creation of multiple word transcriptions. A syntacticstatistical language model that takes into account long-distant word dependencies is proposed for Russian ...
متن کاملPseudo-Syntactic Language Modeling for Disfluent Speech Recognition
Abstract Language models for speech recognition are generally trained on text corpora. Since these corpora do not contain the disfluencies found in natural speech, there is a train/test mismatch when these models are applied to conversational speech. In this work we investigate a language model (LM) designed to model these disfluencies as a syntactic process. By modeling selfcorrections we obta...
متن کاملDetecting Structural Metadata with Decision Trees and Transformation-Based Learning
The regular occurrence of disfluencies is a distinguishing characteristic of spontaneous speech. Detecting and removing such disfluencies can substantially improve the usefulness of spontaneous speech transcripts. This paper presents a system that detects various types of disfluencies and other structural information with cues obtained from lexical and prosodic information sources. Specifically...
متن کاملSynthesising Uncertainty: The Interplay of Vocal Effort and Hesitation Disfluencies
As synthetic voices become more flexible, and conversational systems gain more potential to adapt to the environmental and social situation, the question needs to be examined, how different modifications to the synthetic speech interact with each other and how their specific combinations influence perception. This work investigates how the vocal effort of the synthetic speech together with adde...
متن کاملSpeech disfluency in school-age children's conversational and narrative discourse.
PURPOSE This study was designed to (a) compare the speech fluency of school-age children who do and do not stutter (CWS and CWNS, respectively) within 2 standard diagnostic speaking contexts (conversation and narration) while also controlling for speaking topic, and (b) examine the extent to which children's performance on such discourse tasks is affected by age. METHOD Participants were 44 s...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1996