Abstract

An interactive procedure has been developed to facilitate the assignment and location in the time domain of phonetic symbols to utterances of quantized continuous speech. Conventional methods for transcribing speech data are often based on talker intention and/or listener bias rather than on physical evidence and, generally, symbol location cannot be related to subsequent machine processing. The procedure described operates on-line, in an interactive mode, utilizing a graphic display with an IBM System/360 Model 40. Available for display are a digital sound spectrogram, the power spectrum at a given time sample, an average spectrum for a given sound class, and correlations of a current spectrum with average spectra. With this technique, location and assignment of phonetic symbols are less dependent upon operator expertise (and/or bias) and are highly related to subsequent processing by machine. [Supported in part by Rome Air Development Center.]

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call