Fuji Xerox Co., Ltd.

Matsumi Suzuki

Tetsuro Morino

Saburo Kitamoto

Hiroyuki Kataoka

Toshihiko Toyoshima

Shozo Yokota

Yoichi Ichikawa

Keiichi Ueno

Toshi Itoh

This system verifies the voice of a specific speaker (A) through verification of vocal sounds which form a previously-recorded recognition phrase spoken by A and others. Parallel filters derive a spectral characteristic parameter x.sub.i which contains weighting factors .alpha..sub.i, .beta..sub.i, and which represents a corresponding vocal sound x.sub.i. Improved specificity over other speakers is attained by a Characteristic Feature Extractor which varies .alpha..sub.i, .beta..sub.i, through p different values, thus deriving a set of p parameters (x.sub.ik) for each vocal sound x.sub.i, thus allowing emphasis of the specific vocal sound x.sub.i as spoken by a specific speaker, over that of competing sounds and speakers.

Voice verification system based on word pronunciation

A voice recognition system is disclosed using the centroid of frequency spectra obtained by frequency analysis of a voice as the characteristic parameter of the voice. The locus of the centroid is obtained by periodically measuring, during the utterance of a selected phrase, the average frequency and the average amplitude. The locus constitutes a pattern of the centroid of frequency spectra and is stored in a memory. During recognition a person keys in a known address to extract the stored pattern of his voice from the memory and then reutters the key phrase. The centroid of freqency spectra is recalculated and the locus obtained is compared with the one in the memory. Humming distance is defined as the distance between a recorded locus of centroid of frequency and a newly inputted locus of centroid of frequency.

Voice recognition system using locus of centroid of vocal frequency

Voice verfication of individuals is carried out by utilizing the peak value envelope of voice signals. The voice signal is separated into its positive and negative polarity portions and stepped peak value envelope waveforms are developed from each portion. The latter waveforms are quantized and normalized; the latter being carried out to provide waveforms having a total area therebelow which equals a predetermined area. The quantized, normalized envelopes are then compared with recorded envelopes which were previously calculated in the same manner as the former envelopes. Comparison is carried out by calculating the Hamming distance between the new and stored envelopes.

Speaker identification system using peak value envelop lines of vocal

Speech recognition in prior art uses one extracted characteristic component (x.sub.i) to represent one phoneme (X.sub.i) as spoken by one speaker. This invention provides for recognizing the same phoneme as spoken by different speakers, by deriving a group of such components (x.sub.ik), each a slight variant of the others, to allow finding one component most similar to both the specific phoneme and specific speaker, the method comprising the steps of: normalizing the sound pressure level of an input speech from an unknown speaker; analyzing the normalized voice in a plurality of channels having different frequencies; setting, with respect to the output F.sub.j of each frequency band thus analyzed, a weight .alpha..sub.j of the output F.sub.j so that weight .alpha..sub.j corresponds to a characteristic of a predetermined phoneme X.sub.i ; extracting the characteristic component x.sub.i of the phoneme X.sub.i, setting a weight .beta..sub.j of output F.sub.j so that, when the extracted characteristic component x.sub.i causes a malfunction or error due to another phoneme X.sub.e, a characteristic of phoneme X.sub.e is corresponded to; simultaneously extracting the characteristic component x.sub.e of phoneme X.sub.e and, when the difference between the characteristic components thus extracted is greater than a predetermined threshold value .gamma..sub.i, applying the difference as a characteristic parameter for the phoneme x.sub.i ; expanding the characteristic parameter to obtain a characteristic parameter group based on the characteristic parameter, each being slightly different from each other so as to be adapted for individual characteristics of different speakers; subsequently extracting from the characteristic parameter group a characteristic parameter, having maximum similarity to a reference parameter previously memorized, as an adaptive parameter adaptive to the unknown speaker; and, matching a standard pattern derived from the extracted adaptive parameters with an unknown pattern corresponding to the unknown speakers, thereby effecting recognition or analysis of the voice.

Speech recognition method and apparatus adapted to a plurality of

In a voice verification system, voice parameters which are highly characteristic of particular voices but have only a limited effectiveness life time (such as pitch) are used. The lifetime or similar time indication is recorded along with the original record of the characteristic parameter. At the time of verification, if the lifetime has been exceeded no verification is made. On the other hand if the lifetime of the recorded characteristic parameter has not been exceeded a verification is made. If the verification is positive the newly obtained characteristic parameter is substituted for the originally recorded parameter and the lifetime begins anew.

Method for verifying identity or difference by voice

A half tone reproducing circuitry in a facsimile system characterized by circuitry for pulse number modulating the image signal of an image including half tone circuitry for eliminating those pulses from the pulse frequency modulated signal occurring during the white portion of the image signal and circuitry for on and off controlling an image recording device by the modulated signal thereby reproducing the half tone of the image.

Half tone reproducing process in facsimile

An electromagnetic coupling device for use with a telephone set having a hand hole in the rear thereof, the coupling device comprising a hollow plug removably insertable in the hand hole; a first winding of wire disposed within the hollow plug for receiving electrical signals from the telephone set; a second winding of wire disposed within the hollow plug for transmitting electrical signals to the telephone set, the wire of the second winding being substantially thicker in cross section than the wire of the first winding.

Growing community of inventors

Ebina, Japan

Matsumi Suzuki