Davis, Matt H. (2002) Connectionist modelling of lexical segmentation and vocabulary acquisition. [Book Chapter] (In Press)
Full text available as:
PDF
373Kb |
Abstract
Adults typically hear sentences in their native language as a sequence of separate words and we might therefeore assume, that words in speech are physically separated in the way that they are perceived. However, when listening to an unfamiliar language we no longer experience sequences of discrete words, but rather hear a continuous stream of speech with boundaries separating individual sentences or utterances. Theories of how adult listeners segment the speech stream into words emphasise the role that knowledge of individual words plays in the segmentation of speech. However, since words can not be learnt until the speech stream can be segmented, it seems unlikely that infants will be able to use word recognition to segment connected speech. For this reason, researchers have proposed a variety of strategies and cues that infants could use to identify word boundaries without being able to recognise the words that these boundaries delimit. This chapter, describes some computational simulations proposing ways in which these cues and strategies for the acquisition of lexical segmentation can be integrated with the infants acquisition of the meanings of words. The simulations reported here describe simple computational mechanisms and knowledge sources that may support these different aspects of language acquisition.
Item Type: | Book Chapter |
---|---|
Keywords: | Lexical Segmentation, Vocabulary Acquisition, Simple Recurrent Network, Connectionist Modelling, Language Acquisition, Spoken Word Recognition |
Subjects: | Psychology > Developmental Psychology Neuroscience > Neurolinguistics Psychology > Psycholinguistics |
ID Code: | 2306 |
Deposited By: | Davis, Matthew H. |
Deposited On: | 03 Jul 2002 |
Last Modified: | 11 Mar 2011 08:54 |
References in Article
Select the SEEK icon to attempt to find the referenced article. If it does not appear to be in cogprints you will be forwarded to the paracite service. Poorly formated references will probably not work.
Metadata
- ASCII Citation
- Atom
- BibTeX
- Dublin Core
- EP3 XML
- EPrints Application Profile (experimental)
- EndNote
- HTML Citation
- ID Plus Text Citation
- JSON
- METS
- MODS
- MPEG-21 DIDL
- OpenURL ContextObject
- OpenURL ContextObject in Span
- RDF+N-Triples
- RDF+N3
- RDF+XML
- Refer
- Reference Manager
- Search Data Dump
- Simple Metadata
- YAML
Repository Staff Only: item control page