Abstract: A transcription network having linked computer terminals for a court reporter and for examining, defending, and associate attorneys is disclosed. Cross-reference libraries, phoneme libraries, phoneme dictionaries, and pronunciation indices are used by a reporter terminal to generate readable text transcripts which are communicated to the other terminals for real-time review. Cross-reference libraries provide for the bulk of the transcription while stenographic keystrokes that cannot be directly transcribed are converted to phoneme text or other pronounceable form for down-line readability. A common cross-reference library is provided to supplement local cross-reference libraries to aid transcription and to provide for standardization. The reporter terminal provides for update and preparation of the local cross-reference libraries using case law and evidence prior to a proceeding, as well as provides key-stroke training for the court reporter.

Abstract: A speech coding apparatus and method measures the values of at least first and second different features of an utterance during each of a series of successive time intervals. For each time interval, a feature vector signal has a first component value equal to a first weighted combination of the values of only one feature of the utterance for at least two time intervals. The feature vector signal has a second component value equal to a second weighted combination, different from the first weighted combination, of the values of only one feature of the utterance for at least two time intervals. The resulting feature vector signals for a series of successive time intervals form a coded representation of the utterance. In one embodiment, a first weighted mixture signal has a value equal to a first weighted mixture of the values of the features of the utterance during a single time interval.

Abstract: The system of the invention relates to continuous speech pre-filtering systems for use in discrete and continuous speech recognition computer systems. The speech to be recognized is converted from utterances to frame data sets, which frame data sets are smoothed to generate a smooth frame model over a predetermined number of frames. A resident vocabulary is stored within the computer as clusters of word models which are acoustically similar over a succession of frame periods. A cluster score is generated by the system, which score includes the likelihood of the smooth frames evaluated using a probability model for the cluster against which the smooth frame model is being compared. Cluster sets having cluster scores below a predetermined acoustic threshold are removed from further consideration. The remaining cluster sets are unpacked for determination of a word score for each unpacked word.

Abstract: A vector-quantizing device comprises a first code book for storing a plurality of code vectors at respective addresses; a first addressing unit for selecting a code vector stored in the first code; a synthesis unit supplied with the code vector from the first code book for producing a synthetic signal that simulates an input signal; an evaluation unit supplied with the input signal and further with the synthetic signal from the synthesis means for evaluating a difference therebetween and searching for a code vector that minimizes the difference. There, the evaluation unit produces the first index data that specifies a code vector wherein the difference is minimized as a coded output.

Abstract: A signal processing system for time varying band-limited signals such as speech comprises signal coding means for affording a time encoded signal symbol stream, a matrix generator for generating a fixed size, typically "A" matrix from the symbol stream, and an artificial neural network to which the matrix elements are applied for network training and subsequently for affording an output indicative of the nature of an input signal.

Abstract: A method of coding a speech signal is described. In accordance with the method, a plurality of sets of indexed parameters are generated based on samples of the speech signal. Each set of indexed parameters corresponds to a waveform characterizing the speech signal at a discrete point in time. Parameters of the plurality of sets are grouped based on index value to form a first set of signals which represents the evolution of characterizing waveform shape; the signals of the first set are filtered to remove low frequency components and thereby produce a second set of signals which represents relatively high rates of evolution of characterizing waveform shape. The speech signal is then coded based on the second set of signals representing high rates of characterizing waveform shape evolution. Coding of the speech signal may further be based on a set of smoothed first signals.

Abstract: A voice encoder using a voice activity detector in which two predictive coefficients available from an adaptive predictor in the voice encoder are received for each sample of a input voice signal of the voice encoder. Average values of the predictive coefficients are calculated for each fixed period to decide whether the period is a voice active period or a voice non-active period as a result of comparing the average values with respective ranges of predictive coefficient threshold values predetermined from respective distributions of the two predictive coefficients. Voice active/non-active flags indicative of the voice active period and the voice non-active period are obtained for voice operate switch exchange of encoded of the voice encoder.

Abstract: A post-filter adapted to be used for a speech synthesizing apparatus includes a filtering unit for filtering a synthesized signal, and a scaling coefficient in accordance with both the synthesized signal and a signal output from the filtering unit. The post filter also includes an amplitude detecting unit for detecting an amplitude of the signal output from the filtering unit and for adjusting a value of the scaling coefficient in accordance with a detected result so that an amplitude of the signal output from the filtering unit is kept within a predetermined amplitude value. The post filter further includes a multiplier for calculating the filtering unit with the adjusted scaling coefficient.

Abstract: A voice reproducing device includes a data generator for sequentially providing digital data as an initial value to a recurrence equation and for generating a prescribed number of data with each given an initial value, a decoder for controlling the polarity of its output and the output time thereof in accordance with the data generated by the data generator and for outputting a pulse accordingly, and an analog integrating circuit for integrating the pulse output from the decoder.

Abstract: Apparatus for efficiently encoding a digital signal divides the digital signal into a number of signals within respective predetermined frequency bands which are segmented into blocks which, in turn, are orthogonally transformed so as to form coefficient data. The coefficient data is then quantized. A detection circuit detects whether at least the divided signal within the lowest frequency band contains a transient. If a transient is detected, a control circuit controls the encoding processing of a least the divided signal within the lowest frequency band by, for example, increasing the number of bits used for quantizing, reducing the block size, and/or reducing the signal level so as to reduce the perceptibility of noise in the processed signal.

Abstract: A system for controlling how text is fed to a text-to-speech converter. Text is controllably fed from a text buffer to a text-to-speech converter, the control being based on operator commands which are effectuated at text boundaries, such as word boundaries or sentence boundaries. Operator commands can include commands to advance through the text in a file, in which case text is fed to the text-to-speech converter on a selectable level of granularity, for example, next word, next sentence, next paragraph, etc. It is also possible to advance through the text file rapidly in which case only a sampling of text segments, for example, three words out of twenty, are fed to the text-to-speech converter. Text segments can be repeated, and text segments can be searched for.

Abstract: An audio message storing circuit and a control method thereof in a timer program of a video cassette recorder for recording an audio information simultaneously with the timer program and outputting the stored audio information when the present time is equal to a programmed time. The audio message storing circuit uses a key matrix, a timer, a communication processor, a microphone, a analog/digital converter, a random access memory, a gate and hold circuit, a digital/analog converter, a low band pass filter, an amplifier, and an audio recording circuit coupled to a record/playback head of the video cassette recorder. The method is carried out by the steps of storing the audio information in the random access memory, setting a timer program, and reading the stored information sequentially. The audio signal is outputted to the record/playback head of the recorder when the present time is equal to the programmed time.

Abstract: A recording and reproducing apparatus for recording and reproducing sound using a semiconductor memory as a recording medium, the apparatus including index recording means for storing an index having retrieval information for retrieving an arbitrary position of recorded sound and positional information for indicating a memory address to be retrieved by the retrieval information; and retrieving and reproducing means for retrieving the index containing a specified retrieval information to reproduce the sound from the memory address indicated by the positional information of the index.

Abstract: An apparatus for positively recognizing speech generated by a large, unspecified number of speakers performs optimal speech recognition processing on the basis of speech information dictionaries corresponding to the respective speakers by consecutively inputting and preparing in advance speech information dictionaries corresponding to the respective speakers.

Abstract: An electronic toy which enunciates words or phrases corresponding to sentence parts and a complete sentence by enunciating sounds corresponding to words of phrases selected from a plurality of addresses of a first memory representing audio data corresponding to such words or phrases arranged in each of a plurality of categories. The selected words or phrases correspond to different parts of speech, such as subjects, verbs, adjectives, direct objects and prepositional phrases. The selections are assembled in the second memory and enunciated. The toy may resemble a book in appearance and has printed lettering and color graphics corresponding to the words or phrases.

Abstract: A recognition device and method for recognizing a voice message in the form of pulse code modulation (PCM) digital signals indicative of samples of the voice message. The device and method are adapted such that a recognition result is not provided if the digital signal content satisfies certain requirements which are indicative of a likely erroneous recognition result. The recognition device and method are further adapted to reduce errors in recognizing voice messages of the same message content but different amplitude, as well as to permit simultaneous recognition and storing for recording of a voice message.

Abstract: A record/playback device for use with a removable, interchangeable, flash memory recording medium which enables extended recording comparable with tape cassette dictating equipment. The device includes a housing, a microphone element, control circuitry and a switch mounted on the housing for selecting desired functional operations. A receiving socket is coupled to memory circuitry associated with the control circuitry and is configured for electrical coupling with a flash memory module adapted for receiving and retaining recorded digital information for storage in nonvolatile forme.

Abstract: A GSM mobile telephone includes a digital audio interface for interfacing with a system simulator for test purposes. A digital signal processor within the mobile telephone includes a codec data register that interfaces to an analog codec which provides A/D and D/A conversion of speech signals. An interface circuit is connected to the codec data register and includes a digital audio interface block and a test access port data register, both of which enable test data inputs and test data outputs to be respectively fed to the mobile telephone's digital signal processor and enable test functions to be accomplished without a requirement for loading of test program into the mobile telephone station.

Abstract: A noise signal prediction system includes a signal detector for receiving a mixed signal having a voice signal and a background noise signal and for detecting the presence and absence of the voice signal contained in the mixed signal. A noise level detector is provided for detecting an actual noise level at each sampling cycle during the absence of the voice signal. A storing circuit stores the noise levels for a predetermined number of past sampling cycles. A predicting circuit predicts a noise level of a next sampling cycle based on the stored noise levels in the storing circuit. The storing circuit receiving and stores the actual noise levels during the absence of the voice signal, but stores the predicted noise levels during the presence of the voice signal.

Abstract: A system for matching an input signal, comprising non-white noise and a patterned signal corrupted by said non-white noise, to a plurality of reference signals, the system including an estimator estimating noise features of said non-white noise and producing from the noise features at least one noise whitening filter; a filter generally simultaneously filtering the input signal and the plurality of reference signals using the noise whitening filter and producing a filtered input signal, having a white noise component, and a plurality of filtered reference signals; and a pattern matcher generally robust to white noise for matching the filtered input signal to one of the filtered reference signals.