Abstract: The present invention is directed to detecting information such as characters or the like formed on a form using fluorescent ink without using any special components such as a light source for emitting ultraviolet light, an optical filter for intercepting transmission of ultraviolet light, or the like. When visible light which is emitted by fluorescent ink excited upon being irradiated with ultraviolet rays is detected by an image sensor, a level of the output photoelectrically converted by the image sensor is saturated and assures a level higher than white level as white reference, so that information such as characters or the like formed on a form using fluorescent ink can be acquired.

Abstract: An utterance boundary detecting apparatus of this invention includes an acoustic processor for generating speech parameter time sequence data according to an input speech signal. The speech parameter time sequence data generated from the acoustic processor is delivered to a buffer memory and noise level determining circuit. The noise level determining circuit calculates the average value of speech parameter values of a background noise corresponding to a silent period when a speech signal is input as words uttered. The apparatus includes a threshold value calculating circuit for calculating an utterance boundary detection threshold value on the basis of an average value calculated by the noise level determining circuit. An utterance boundary detecting circuit generates utterance boundary data on the basis of the threshold value from the threshold value calculating circuit and speech parameter time sequence data in a buffer memory.

Abstract: An acoustic signal processing circuit extracts input speech pattern data and subsidiary feature data from an input speech signal. The input speech pattern data comprise frequency spectra, whereas the subsidiary feature data comprise phoneme and acoustic features. These data are then stored in a data buffer memory. The similarity measures between the input speech pattern data stored in the data buffer memory and reference speech pattern data stored in a dictionary memory are computed by a similarity computation circuit. When the largest similarity measure exceeds a first threshold value and when the difference between the largest similarity measure and the second largest measure exceeds a second threshold value, category data of the reference pattern which gives the largest similarity measure is produced by a control circuit to correspond to an input speech.

Abstract: The detection of voice (speech) signal presence in input signal-plus-noise is improved by more accurate determination of the decision threshold, which is determined by first finding a medium-length interval consisting of noise-signal-noise (no-signal, signal, no-signal), then calculating a histogram (energy probability distribution) for the interval, then finding the maximum value of variance of the histogram as the optimal threshold, plus an arbitrary offset.