Phoneme recognition using visual features on speech spectrograms. Katagiri, S and Yokota, M In European Conference on Speech Technology, pages 1365-1368. Edinburgh, Scotland, UK. September 1987.
Phoneme recognition using visual features on speech spectrograms [link]Paper  abstract   bibtex   
In order to apply speech spectrogram reading heuristics to an automatic speech recognition system, a more accurate expression of the heuristics must be developed. In particular, the transformation between acoustic feature measurements and phoneme candidates must be developed in a quantitative manner. In this paper, a visual acoustic-feature label and a phoneme identification approach using this label is proposed. The visual acoustic-feature label, which is a polygon on a speech spectrogram, represents some aspects of an acoustic feature by its own geometric characteristics. Preliminary experimental results show that phoneme identification using the visual acoustic-feature label is feasible for realizing the quantitative transformation rules between the acoustic feature measurements and phoneme candidates.
@incollection{katagiri_phoneme_1987,
	Author = {Katagiri, S and Yokota, M},
	Booktitle = {European Conference on Speech Technology},
	Date = {1987},
	Date-Modified = {2016-09-24 18:56:06 +0000},
	File = {Attachment:files/5836/Katagiri, Yokota - 1987 - Phoneme recognition using visual features on speech spectrograms.pdf:application/pdf},
	Keywords = {acoustic phonetics, phonetics, speech technology},
	Pages = {1365-1368},
	Publisher = {Edinburgh, Scotland, UK. September 1987},
	Title = {Phoneme recognition using visual features on speech spectrograms},
	Url = {http://www.isca-speech.org/archive/ecst_1987/e87_1365.html},
	Abstract = {In order to apply speech spectrogram reading heuristics to an automatic speech recognition system, a more accurate expression of the heuristics must be developed. In particular, the transformation between acoustic feature measurements and phoneme candidates must be developed in a quantitative manner. In this paper, a visual acoustic-feature label and a phoneme identification approach using this label is proposed. The visual acoustic-feature label, which is a polygon on a speech spectrogram, represents some aspects of an acoustic feature by its own geometric characteristics. Preliminary experimental results show that phoneme identification using the visual acoustic-feature label is feasible for realizing the quantitative transformation rules between the acoustic feature measurements and phoneme candidates.},
	Bdsk-File-1 = {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},
	Bdsk-Url-1 = {http://www.isca-speech.org/archive/ecst_1987/e87_1365.html}}
Downloads: 0