Rule‐controlled data base search

Abstract

In current acoustic‐phonetic research, there is a need for huge speech data bases. There are considerable problems in administering such data bases, both to transcribe and segment the speech and to easily access stored material. We have created a speech analysis system to attempt to alleviate these problems. Speech data are stored in sentence‐sized files. These files are segmented and transcribed semiautomatically [M. Blomberg and K. Elenius, STL/QPSR 1/(1985)] given a phonetic transcription of the utterance. This transcription is generated by the text‐to‐phonetic component of our synthesis by rule system [R. Carlson and B. Granstrom, IEEE‐ICASSP (1976)]. The same rule structure, similar to the notation used in generative phonology, is used for accessing the data. By a brief rule statement, speech segments meeting the specified contextual conditions can be identified. Durational data can be collected directly during the data base search. Spectral analysis programs, operating with a variety of spectral representations [R. Carlson and B. Granstrom, Eds., The Representation of Speech in the Peripheral Auditory System (Elsevier/North Holland)] have also been created that display the result, typically as a mean/ s.d. spectrum or contour histogram spectrum.