Third International Conference on Spoken Language Processing (ICSLP 94)

Yokohama, Japan
September 18-22, 1994

Automatic Context-Sensitive Measurement of the Acoustic Correlates of Distinctive Features at Landmarks

Mark Johnson

Research Laboratory of Electronics, MIT, Cambridge, MA, USA

This paper models speech recognition as the estimation of distinctive feature values at articulatory landmarks [8]. Toward this end, we propose modeling each distinctive feature as a table containing phonetic contexts, a list of signal measurements (acoustic correlates) which provide information about the feature in each context, and, for each context, a statistical model for evaluating the feature given the measurements. The model of a distinctive feature may include several sets of acoustic correlates, each indexed by a different set of context features. Context features are typically lower-level features of the same segment, e.g. manner features ([continuant, sonorant]) provide context for the identification of articulator-bound features ([lips, blade]). The acoustic correlates of a feature can be any static or dynamic spectral measurements defined relative to the time of the landmark. The statistical model is a simple N-dimensional Gaussian hypothesis test. A measurement program has been developed to test the usefulness of user-defined acoustic correlates in user-defined phonetic contexts. Measures of voice onset time and formant locus classification are presented as examples.

Full Paper

Bibliographic reference.  Johnson, Mark (1994): "Automatic context-sensitive measurement of the acoustic correlates of distinctive features at landmarks", In ICSLP-1994, 1639-1642.