Abstract: An encoder apparatus for communication and pattern recognition systems employing a nearest neighbor search method for signal and data compression, based on a vector encoding quantization technique which optimizes systems performance with substantially reduced computational complexity using a fast geometrically-oriented search procedure. The apparatus comprises pre-procesing apparatus for providing off-line reorganization of a codebook having a set of reference vector patterns constituting codevectors with which the input vector is to be compared in a search procedure, and on-line apparatus for encoding the random input vector through quantization in accordance with the search procedure in the codebook.

Abstract: A sensory type pattern such as a speech or other sound pattern is analyzed to obtain the spectral distribution of the neural response thereto. A plurality of logarithmically related neural response intensity threshold signals is formed. The frequency spectrum of the sensory type pattern is divided into a plurality of overlapping spectral portions and the waveform of each prescribed spectral portion is partitioned into successive time segments. For the current time segment of each spectral portion waveform, the time intervals between crossings of the neural response intensity threshold level signals by the spectral portion waveform are detected and signals representative of the counts of inverse time intervals between the crossings of the plurality of levels are generated to form an inverse time interval histogram for the spectral portion.

Abstract: The present invention describes a method and arrangement for reducing a sequence of initial frames into a reduced set of representative frames by combining the initial frames into a plurality of representative frames, the combining process including generating a distortion measure associated with each representative frame and comparing each distortion measure to a distortion threshold. From these representative frames, a set of mutually exclusive frames is determined to minimize the number of representative frames, whereby each representative frame in the set represents a unique set of contiguous initial frames and has an associated distortion measure which does not exceed the distortion threshold.

Abstract: A semiconductor read only memory (ROM) serves to store digitized audio signals in a redundancy free form. The output of the memory is coupled to the input of a data expander in the playback circuit. The memory may also store an algorithm which serves to control operation of the data expander before the signals are further processed via a digital-to-analog converter and where the output of the digital/analog converter is then coupled to the input of an analog amplifier and then to a loud speaker arrangement. In this manner, digitized audio signals can be operated on by either error correcting modules or data expansion modules, both of which modules may include microprocessors where the algorithms for control of the microprocessors are also stored within the read only memory. In this manner, all mechanical movements between the storage medium and the pick up devices, which are typical of the prior art, is eliminated as the main storage medium consists of a read only memory or a plurality of such devices.

Abstract: Multiple similarities between phoneme feature data extracted from a word input signal and reference pattern data read out from a reference pattern memory are calculated by a similarity calculation section. The multiple similarities are converted by a conversion filter into posterior probabilities using parameter sets which are read out from a filter parameter bank in accordance with the obtained multiple similarities. The posterior probabilities are subjected to matching with data stored in word dictionary by a word matching section, thus discriminating the input word.

Abstract: In an encoder for encoding a speech signal having a spectrum envelope into a plurality of excitation pulses, a spectrum emphasis unit emphasizes peak components of the spectrum envelope to produce an emphasized speech signal. As a result of a spectrum emphasis operation, the emphasized speech signal has an emphasized spectrum envelope which substantially comprises a plurality of line spectra. Responsive to the emphasized speech signal, a pulse producing unit produces a plurality of excitation pulses by the use of a pulse search method.

Abstract: In a digital system for amplitude modulation, the logical signal at acoustic frequency is modulated by means of an ultrasonic frequency signal, the duty-cycle of which can be varied by means of a program. More particularly, the modulating signal is generated by a cyclic counter. The duty-cycle is programmed by writing the corresponding numerical value to a register with a capacity equal to that of the counter. An adder, adding the value of the counter each time to that of the register, generates a carry signal which gives consent to the acoustic signal only for a part of the relative cycle of the modulating signal.

Abstract: For use in deciding optimum pole and zero parameters for an input signal with reference to an input cepstrum of the signal, candidate pole and zero parameters are stored in advance in a table. Supplied with an impulse and controlled by a candidate set pair of a pole parameter set and a zero parameter set selected from the candidate pole and zero parameters of the table, first and second filters produce first and second outputs defined by terms of up to a certain order. Responsive to the first and second outputs and to factors of multiplication which are given by inverse numbers of time intervals related to the respective terms, an analysis filter produces a converted signal which is equivalent to a model output cepstrum of a model output signal produced by a pole-zero model defined by the candidate set pair. A cepstrum subtracter calculates a cepstrum difference between the input cepstrum and the converted signal.

Abstract: A system for implementing a bus window interlock scheme between a first and a second bus utilizes two bus window modules. The first bus window module coupled to the processor bus includes an interlock state bit which is set upon the acceptance of an interlock transaction from a processor. No further interlock transactions will be accepted while the interlock state bit is set. The interlock transaction is passed to a transaction buffer in the second bus window module which is coupled to memory through the memory bus. The transaction buffer passes the interlock data for memory to the memory bus while simultaneously loading a one deep interlock buffer. A confirmation is sent from the memory back to the transaction buffer. If the confirmation is interlock busy, then the interlock transaction is retried from the interlock buffer thus allowing the transaction buffer to process other commands.

Abstract: An improved excitation vector generation and search technique (FIG. 1) is described for a code-excited linear prediction (CELP) speech coder (100) using a codebook memory of excitation code vectors. A set of M basis vectors v.sub.m (n) are used along with the excitation signal codewords (i) to generate the codebook of excitation vectors u.sub.i (n) according to a "vector sum" technique (120) of converting stored selector codewords into a plurality of interim data signals, multiplying the set of M basis vectors by the interim data signals, and summing the resultant vectors to produce the set of 2.sup.M codebook vectors. Only M basis vectors need to be stored in memory (114), as opposed to all 2.sup.M code vectors.

Abstract: A system for subband coding of a digital audio signal x(k) includes in the coder (1) a filter bank (3) for splitting the audio signal band, with sampling rate reduction, into subbands (p=1, . . . P) of approximately critical bandwidth and in the decoder (2) a filter bank (5) for merging these subbands, with sampling rate increase. For each subband (p) the coder (1) comprises a detector (7(p)) for determining a parameter G(p;m) representative of the signal level in a block (p;m) of M samples of the subband signal x.sub.p (k) as well as a quantizer (8(p)) for adaptively block quantizing this subband signal in response to parameter G(p;m), and the decoder (2) comprises a dequantizer (9(p)) for adaptively block dequantizing the quantized subband signal s.sub.p (k) in response to parameter G(p;m).

Abstract: An apparatus for calculating a filter factor for a digital filter which can be simply constructed and in which the time period for calculation can be shortened as compared with conventional techniques. The apparatus has an input circuit for inputting a desirable frequency property, a division circuit for dividing the inputted frequency property into a plurality of frequency bands, and a calculating circuit for obtaining filter factors for realizing a frequency property of each of the divided frequency bands. The input circuit inputs the frequency property with a frequency resolution corresponding to the number of the filter factors. The division circuit performs a correction for a division so that the frequency property becomes zero from a frequency over the high cut-off frequency of a band-pass filter toward the Nyquist frequency of the frequency band corresponding to the transversal filter, the Nyquist frequency being 1/2 of the sampling frequency.

Abstract: A method for specifying and executing independent, multi-media tasks along a synchronizing time-line, preferably in the form of a spreadsheet matrix with the event elements making up the rows and the time periods the columns. The media include various pieces of hardware such as touch screens, voice synthesizers, video disk players, and the like. In designing a presentation, the author types into the matrix indications of which piece of hardware will be operating and for what specified period of time during the presentation. This information is also accessible to all other events in the presentation thereby allowing complex multi-media presentations to be designed by a user who is relatively unsophisticated in using computers.

Abstract: In a mobile communications system, a plurality of base stations may receive the same signal, whereby the "best" reception may be determined. The plurality of received digital signals are combined to form a majority signal. Each of the received digital signals is compared to the majority signal to estimate the bit-error-rate (BER) of each received signal. If the difference between the highest and lowest BER is less than a predetermined threshold, the majority signal is selected for reception. Conversely, if the BER difference is greater than the predetermined threshold, the received digital signal with the lowest BER is selected for reception. Thus, the signal with the lowest BER is selected to provide the best possible communications.

Abstract: A printer which receives data and commands which are transferred from a data processing apparatus serving as an external apparatus such as a host computer and prints the data. This printer includes a setting circuit to set information regarding the printing; this setting circuit having a first mode in which the information regarding the printing can be designated irrespective of the information from the external apparatus and a second mode in which the information regarding the printing is designated on the basis of the information from the external apparatus. A deciding circuit to decide whether the information regarding the printing has been set to predetermined information by the setting circuit or not and a control unit for controlling the printer in a manner such that the priority is given to the first mode when the information regarding the printing has been set to information other than the predetermined information on the basis of the decision of the deciding means.

Abstract: A multiple window display system includes a display device and a screen ownership area pointing to the identity of the window which is to contribute the data for each display area of the display device. An ordered list is maintained of the active windows in the priority order thereof. Means are provided to regenerate the screen ownership area from the ordered list, on each change made to the list, in terms of list position per device display area, by overwriting, progressing through the list in order of increasingly significantly priority order, the list indicating, in each position thereof, the identity of the window having the respective priority. The list contains the addresses of the windows in storage and the type thereof. The screen ownership area is reset to the lowest potential priority list position value and is overwritten.

Abstract: Upon the adjustment of a control setting of a test instrument, the current data (taken before the change is effected) is manipulated by recalculating the trace data in light of any adjustment that is entered by a user. This represents an estimate of the effect of the adjustment in the control setting, and this estimate is redrawn on a display device. This modified image is preferably displayed before the real new image is ready to be displayed. This provides immediate feedback to the user so that adjustments can be effected without slowing the measurement process. Various embodiments provide panning or repositioning the trace left and right while changing the x-axis center frequency; stretching or compressing the trace (spanning) while changing the frequency span; or using a combination for adjustments to only the start or stop frequency. In one embodiment, the estimate can be displayed only after the user has committed to having a new measurement performed.

Abstract: A signal processing system for a spectrum analyzer presents a three-dimensional display of a signal in the form of a display of an independent variable, such as frequency, versus a selected property of the signal, such as power, where the selected property is represented as a distribution of values for each independent variable interval in the form of intensity variation on the display. The independent variable is divided into m slots, and the selected property is sampled k times for each slot. A histogram for each slot is built up in a buffer representing the number of samples occurring at each of n power levels. The histogram is transferred to a display memory column by column for each slot, and after all the slots have been processed the display memory is read out onto a display device, such as a CRT, in a raster scan format with the intensity of each pixel element being a function of the number or samples at that power level within that frequency interval.

Abstract: Speech is data-compressed and coded for transmission using TDHC (Time Domain Harmonic Compression) for the voiced signal, and decimated sampling for the unvoiced signal. Features include voiced/unvoiced detection, pitch period detection, border detection, coding by ADPCM, and optimum quantization coding. Reception involves decoding and data-reconstruction.

Abstract: A computer speech system for digitally storing and reproducing representations of human speech. An analog waveform representative of a segment of speech is compressed for storage by storing companded differences between adjacent local maxima and minima of the waveform together with the lengths of time between the occurrences of the maxima and minima. The compressed speech is reproduced by looking up precomputed values in a lookup table according to the companded differences and times and furnishing the values from the lookup table to a digital-to-analog converter and thence to a conventional audio output device. The values are furnished to the converter by putting them on lower bits of the address bus and performing an operation at an address to which the converter responds.