Learn More
A speaker's regional dialect is a rich source of information about that person. Two studies examined five- to six-year-old children's perception of regional dialect: Can they perceive differences among dialects? Have they made meaningful social connections to specific dialects? Experiment 1 asked children to categorize speakers into groups based on their(More)
Learning to group words into phrases without supervision is a hard task for NLP systems , but infants routinely accomplish it. We hypothesize that infants use acoustic cues to prosody, which NLP systems typically ignore. To evaluate the utility of prosodic information for phrase discovery, we present an HMM-based unsupervised chunker that learns from only(More)
A well-known effect in speech production is that more predictable linguistic constructions tend to be reduced. Recent work has interpreted this effect in an information-theoretic framework, proposing that such predictability effects reflect a tendency towards communicative efficiency. However, others have argued that these effects are, in the terminology of(More)
—Documents from the same domain usually discuss similar topics in a similar order. In this paper we present new ordering-based topic models that use generalised Mallows models to capture this regularity to constrain topic assignments. Specifically, these new models assume that there is a canonical topic ordering shared amongst documents from the same(More)
Stress is a useful cue for English word segmentation. A wide range of computational models have found that stress cues enable a 2-10% improvement in segmen-tation accuracy, depending on the kind of model, by using input that has been annotated with stress using a pronouncing dictionary. However, stress is neither invariably produced nor unambiguously(More)
A well-known effect in speech production is that more predictable words tend to be phonetically reduced. Recent work has suggested that predictability effects result from hardwired properties of the language production system, rather than active modulation by the talker to accommodate the listener. However, these studies investigated only minor(More)
1. Background • Listeners exhibit modest improvement in classifying unfamiliar talkers by regional dialect of American English when trained on sentence-length utterances with feedback (Clopper and Pisoni, 2004). • Listeners learn to identify unfamiliar talkers by name more slowly with word-length utterances than sentence-length utterances (Nygaard and(More)
This course is an introduction to computational linguistics. The book for the course is Speech and Language Processing (J&M, 2008), and we'll also use parts of the Mackay book " Information Theory, Inference, and Learning Algorithms " (Mackay, 2003), available on-line for free at (or for purchase as a hardcopy). There will be two short projects.