Abstract: A signal pattern recognition system includes a reference pattern memory storing plural reference data patterns against which input patterns are compared for recognition. The reference data patterns are formed by merging training patterns together. Each training pattern, to be accepted for merging, must match the previously merged patterns by a threshold amount. The threshold is automatically varied as the number of previously merged training patterns increases. If a predetermined number of successive training patterns is below the threshold, the training process is repeated, from the beginning, for the reference pattern which generates the errors. The system automatically trains each reference pattern with the same number of training patterns to assure uniformity when input data patterns are compared for recognition.

Abstract: A speech synthesizer for developing a signal representing both voiced and fricative sounds from signals representing blocks of externally generated data, the synthesizer including a period counter for delineating a series of pitch periods each having a duration specified by the data, a source generator for developing an excitation signal having an amplitude specified by the data, the excitation signal being developed at the beginning of each of a series of constant duration sample intervals each with a random sign for fricative sounds, the excitation signal being developed at the beginning of each of the pitch periods for voiced sounds and a time multiplexed digital filter having characteristics specified by the data including circuit switching to control whether the filter functions as a second-order filter or a cascaded pair of first-order filters, the filter being driven by the excitation signal during the beginning of each of the sample intervals and by the output of the filter thereafter so as to develop

Abstract: The output of an audio power amplifier fed to a loudspeaker driver is sensed and when this output exceeds a predetermined level, a switch control circuit operates to place a current limiting device, such as a power resistor, in series with one of the driver feed lines, thereby reducing the power fed thereto. An indicator may also be provided to indicate when the overload condition exists. Operation is made more accurate and distortion minimized by employing separate sensing and power channels in the circuit, the first of these channels operating to sense the overload while the second of these channels utilizes a portion of the amplifier output for driving the switch control circuit for placing the power limiter in the speaker line. The sensing channel is designed to sense the average voltage of the amplifier output, thus providing control more directly responsive to power. The power channel is designed to use audio power from the amplifier output only occasionally, thus producing a minimum of distortion.

Abstract: This communication system reduces required data transmission channel capacity by encoding source words into encoded words with usually less binary bits for transmission. Features of the invention include a hashing algorithm which transforms the source words into psuedorandom code words used for both transmission and addresses in a vocabulary memory. Before transmission each source word and its hash code is checked against a pre-recorded vocabulary memory to decide which should be transmitted.

Abstract: Voice pitch is detected by passing a speech signal from a microphone 10 through a filter array 20 to a summing amplifier 22. The summing amplifier 22 provides a composite signal of the acoustic components of the speech signal and this composite signal is input to a multiplier 26 and from the multiplier 26 to a multiplier 36 having an output connected to a comparator 42. The comparator 42 compares the output of the multiplier 36 with a hysteresis signal, then provides an output to a phase locked loop 46. The phase locked loop 46 provides a first voltage output in analog form to a bar graph display 50 and provides a square wave output that is applied to a frequency-to-voltage converter 52 for driving a digital display 54.

Abstract: An improved method of verifying the speaker from whom long term spectral characteristics have been derived during a learning phase. Mean values and distance thresholds have been calculated and stored. Improved recognition is obtained by incorporating into the stored values additional values of speech samples in ranges situated outside the distance threshold value. When, during verification, a speech sample not situated within the distance threshold value is encountered its distance from the additionally stored speech samples is successively determined and compared with the separate distance threshold value. The speaker is rejected only if the speech sample to be verified is not situated at a sufficiently small distance from any of the additionally stored speech samples. The method thus increases the recognition range to better approximate the actual distribution of speech samples of the learning phase.

Abstract: A signal encoder and classifier particularly adapted to speech recognition includes a circularly addressed buffer which is independently addressed by a new data writing address system and a buffered data reading system so that writing and reading of data may be accomplished on a time shared basis. This time shared operation permits serial writing and reading of the pattern data without interrupting income signal storage. The writing data address system addresses the data into the buffer in a circular fashion while the reading data address system utilizes stored addresses identifying the beginning and end of the signal patterns for addressing sequential patterns from the buffer.

Abstract: Effective expansion of a common intermediate buffer memory by equivalent use of the buffer memory in each CPU in a multiprocessor system. A method and system for achieving buffer memory coincidence is applied to a multiprocessor system provided with central processing units, buffer memories contained in respective central processing units, a main memory, and an intermediate buffer memory connected between the main memory and the buffer memories, wherein a buffer invalidation address information (BIA GO) is sent from the intermediate buffer memory to the i-th central processing unit (BIA GO #i) in accordance with the following logical expression:BIA GO #i=REQ.CPU.times.W.times.{F.upsilon.F.times.(COPY#i.upsilon.VIF)}where the term "REQ.

Abstract: A tone control arrangement is disclosed which is suitable for use in various sound reproducing instruments and in which the signal-to-noise ratio of the reproduced sound can be improved by controlling the frequency response of the audible signal with a variable d.c. voltage source. Frequency response control by a single d.c. control voltage is achieved by connecting in series and in order a first mixer, passive circuit network having a signal transmission characteristic determined by the desired frequency response of the output sound signal to be reproduced, a multiplier and a second mixer. The inverting inputs of the first and second mixers both receive the output of the multiplier, while the non-inverting input of the first mixer receives the input audio signal. The non-inverting input of the second mixer receives either the output of the first mixer or the input audio signal. Finally, the second input to the multiplier is provided by a variable d.c. voltage source.

Abstract: A microphone which uses noise-adaptive automatic output level control in speech recording. An input amplifier cooperates with a voltage doubler to form an automatic gain control. A diode and capacitor circuit provides for envelope detection of the signal which is then differentiated, amplified and detected to result in a DC control signal. This control signal is applied to an output amplifier which functions as a three pole Chebyshev filter in which a reduction of gain dampens one pole of the filter causing the response to drop off, so that the signal appearing at the output of the microphone has a substantially high speech-to-noise ratio. The output signal appears at a constant level where it may then be acted upon by the automatic gain control of an associated tape recorder connected to the microphone.

Abstract: A continuous speech recognition system utilizes a format memory (14) which specifies a sequence of word sets and a plurality of words, or reference patterns, which may be included in each word set. The input pattern sequence is divided into all possible partial patterns having start points p and end points q, and each of these partial patterns is compared with all reference patterns to derive elementary similarity measures. The elementary similarity measures for each combination of a partial pattern and a permitted word in a word set under the specified format are then examined to determine the optimum input pattern segmentation points and corresponding sequence of reference patterns which will yield a maximum similarity result. The maximum similarity is represented by ##EQU1## where S(p(x-1), p(x),n(x)) indicates the degree of similarity between an input partial pattern having a start point p(x-1) and an n point p(x) and a reference word unit n(x) within a word set f.sub.

Abstract: An audiometric testing system containing a processor and programmable analog circuitry for controlling frequency and intensity of audiometric test tones communicates with a remote data processing system via a communications link. The remote data processing system stores prior audiometric records for a large number of subjects. Each day, the audiometric records of a number of selected persons are transmitted from the remote data processing system to a memory of the audiometric testing system in response to identification numbers entered by an operator. The audiometric responses of each person to patterns of sounds of varying frequencies and intensities are determined. The varying frequencies and intensities are determined by a bracketing subroutine of an algorithm stored in and executed by a processor of the audiometric testing system. A person's responses to such test tones are utilized by the bracketing subroutine to determine his threshold levels at the respective frequencies for each ear.

Abstract: Sounds produced by human or mechanical sources are recognized by comparison of certain waveform characteristic ratios with prestored ratios. Transitions and transition glides of characteristics are used to select certain stored groups of ratios for comparing.

Abstract: Signal-to-noise is enhanced by including the upper sideband of a sampled-speech signal with the original baseband signal for further signal processing. The invention features shifting both bands to form a continuum from 0 Hz to the sampling frequency. Application in a speech recognition is shown.

Abstract: A synthetic speech calculator is provided which is capable of generating an audible sound signal which can be representative of keyed data, sequences of keyed data (and errors), processed data, the contents of a memory or the contents of a counter. The sound signal can be suppressed under certain specific circumstances or made soft in intensity. A synthetic-speech calculator includes the keyboard consisting of digit keys and function keys, a desired number of registers for storing information entered by the depression of selected ones of the digit keys, a read-only-memory for storing a large number of digital codes as sound quantizing information, counter means for specifying the address of the memory so as to take a specific digital code out of the memory, a digital-to-analog converter for converting the specific digital code taken out of the memory into an audible sound signal, and a loud speaker driven by the audible sound signal and producing an audible sound.

Abstract: A speech analyzer with improved pitch period extraction and improved accuracy of voiced/unvoiced decision comprises circuits for calculating autocorrelation coefficients forwardly and backwardly with respect to time. Reference members for the forward and the backward calculation are those successively prescribed ones of windowed samples of a signal representative of speech sound which are placed in each window period farther from a trailing and a leading end thereof, respectively. Members to be joined to the respective reference members for forward and backward calculation of each autocorrelation coefficient are displaced therefrom by a joining interval farther from the leading and the trailing ends, respectively. The joining interval is varied between a shortest and a longest pitch period of the speech sound stepwise by a spacing between two successive windowed samples.

Abstract: A system for detecting the pitch of a voice signal, in which a plurality of pitch searching periods are determined so that pitch components of multiple relationship are not included in each of the pitch searching periods, and in which after detecting a pitch searching period including the pitch from the pitch searching periods, the pitch searching periods are adaptively shifted in a mannger to follow the change direction of the pitch predicted from the result of detection of the detected pitch.

Abstract: In a pattern recognition device according to pattern matching, one or more specific dimensions of vector components are memorized for each reference pattern feature vector sequence in a reference pattern memory for the reference pattern feature vector sequences. A warping function for time-normalizing input pattern feature vectors of a sequence and the vectors of each reference pattern feature vector sequence is determined so as to minimize the difference between a pattern represented by the specific vector components of the specific dimension or dimensions and another pattern represented by the vector components corresponding in the input pattern feature vector sequence to the specific reference pattern feature vector components as regards the dimensions of a space in which each input or reference pattern feature vector is defined. The input pattern feature vector sequence and each reference pattern feature vector sequence are subjected to nonlinear pattern matching with reference to the warping function.

Abstract: The classification of a signal on a telephone line can be based on its short term average energy level; nonspeech (data or signalling) as high level; speech as medium level, and noise as low level. A common use for speech detectors occurs with a time assignment speech interpolation (TASI) system. A speech detector extends a trunk activity signal (TAS) for indicating whether energy or noise is detected on a trunk. Responsive to the TAS signal, a processor processes a signal on the trunk. On occasion, it may be desirable to process different energy signals according to different methodologies. For example, during overload, a TASI system may truncate bits of a speech signal and not truncate bits of a nonspeech signal. For classifying energy signals as between speech signals and nonspeech signals, the instant classifier includes an arrangement for extracting at least one feature, or property, such as the short term power of the signal to be classified.

Abstract: Upon analyzing grammatically and phonetically a printed text for accents, pauses, intonations and influences of adjacent voice elements in a sentence to be synthesized, a computer loads a plurality of registers including an address counter with instructions for addressing a read-only memory, these instructions specifying rates of counting, numbers or counts, whether counting is to be decremental or incremental and initial addresses of sequences of binary bits coding successive magnitudes of noise signals or of voice-frequency functions. The output of the read-only memory is fed to a loudspeaker via a digital/analog converter and an amplifier whose output is modulated by a signal transmitted from the computer through another d/a converter. The durations of noise and voice-frequency speech elements read out from the memory and the modulation of their amplitudes by the amplifier are randomly modified within .+-.3% for the frequency and .+-.