Learning the Hidden Structure of Speech.

Autor/inn/en	Elman, Jeffery Locke; Zipser, David
Institution	California Univ., San Diego, La Jolla. Inst. for Cognitive Science.
Titel	Learning the Hidden Structure of Speech.
Quelle	(1987), (28 Seiten)Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; Monographie
Schlagwörter	Articulation (Speech); Artificial Speech; Communication Research; Computers; Distinctive Features (Language); Learning Processes; Phonetics; Recognition (Psychology); Research Methodology + Suchen Sie Ihr Suchwort? Kommunikationsforschung; Digitalrechner; Learning process; Lernprozess; Phonetik; Fonetik; Recognition; Wiedererkennen; Research method; Forschungsmethode
Abstract	The back-propagation neural network learning procedure was applied to the analysis and recognition of speech. Because this learning procedure requires only examples of input-output pairs, it is not necessary to provide it with any initial description of speech features. Rather, the network develops on its own set of representational features during the course of learning. A series of computer simulation studies were conducted to assess the ability of these networks to label sounds accurately, to learn to recognize sounds without labels, and to learn feature representations of continuous speech. The studies demonstrated that the networks can learn to label presegmented naive sound tokens with accuracies of up to 95%. Networks trained on segmented sounds using a strategy that requires no external labels were able to recognize and delineate sounds in continuous speech. These networks developed rich internal representations that included units that corresponded to such traditional distinctions as vowels and consonants, as well as units that were sensitive to novel and nonstandard features. Networks trained on a large corpus of unsegmented, continuous speech without labels also developed interesting feature representations that may be useful in both segmentation and label learning. The results of the studies, while preliminary, demonstrate that back-propagation learning can be used with complex, natural data to identify a feature structure that can serve as the basis for both analysis and nontrivial pattern recognition. (Author/FL)
Erfasst von	ERIC (Education Resources Information Center), Washington, DC