Abstract: An analyzing unit 1 converts an input speech into a feature vector time series. A reference pattern storing unit 3 stores the feature vector time series obtained by the same manner as in the analyzing unit. A matching unit 2 correlates for time axis the input speech feature vector time series and the reference patterns to one another. An environmental adapting unit 4 performs the environmental adaptation between the input speech feature vector time series and the reference patterns according to the result of matching in the matching unit 2. A speaker adapting unit 6 performs the adaptation concerning the speaker between the environmentally adapted reference patterns from the environmental adapting unit 4 and the input speech feature vector time series.

Abstract: A speech detection system is provided with multiple speech detector sub-systems. The speech detection sub-systems employ distinct statistical methods for determining whether speech is present in an electronic communication signal received at an output terminal. For example, a first speech detection sub-system employs moving average peak signal filter, a second speech detection sub-system employing a moving average noise filter, and a third speech detection sub-system employs a variance filter. Signals from each of the filters are compared with respective threshold values, and the threshold values are provided to speech determination logic for making an aggregate speech detection decision. The speech detection system is useful for telephonic automatic gain control.

Abstract: A method and system for editing words that have been misrecognized. The system allows a speaker to specify a number of alternative words to be displayed in a correction window by resizing the correction window. The system also displays the words in the correction window in alphabetical order. A preferred system eliminates the possibility, when a misrecognized word is respoken, that the respoken utterance will be again recognized as the same misrecognized word. The system, when operating with a word processor, allows the speaker to specify the amount of speech that is buffered before transferring to the word processor.

Abstract: Provided is an improved graphical user interface method and system to be utilized with voice input. The method and system remove a voice activated commands dialog box when a voice input system is not active. The method and system achieve the forgoing objects by (1) sensing whether a voice input system is active or inactive, which can be achieved by determining whether a voice input program is set to actively receive voice input in response to a change of state of a voice input device (such as a microphone); and (2) removing a voice activated commands dialog box from prominent display within a graphical user interface if the voice input system is sensed inactive.

Abstract: In a voice coding and decoding method and apparatus using an RCELP technique, a CELP-series decoder can be obtained at a low transmission rate. A voice spectrum is extracted by performing a short-term linear prediction on voice signal. An error range in a formant region is widened during adaptive and renewal codebook search by passing said preprocessed voice through a formant weighting filter and widening an error range in a pitch on-set region by passing the same through a voice synthesis filter and a harmonic noise shaping filter. An adaptive codebook is searched using an open-loop pitch extracted on the basis of the residual minus of a speech. A renewal excited codebook produced from an adaptive codebook excited signal is searched. Finally, a predetermined bit is allocated to various parameters to form a bit stream.

Abstract: An audio recording and reproducing apparatus includes a controller for controlling the entire behaviors, hard disc for write and read of audio data, audio compression/expansion circuit for expanding compressed audio data, and external I/O port. The audio recording and reproducing apparatus is connected to a network service center to obtain desired music data from storage of the network service center and to store it in the hard disc.

Abstract: An automated speech recognition system converts a speech signal into a compact, coded representation that correlates to a speech phoneme set. A number of different neural network pattern matching schemes may be used to perform the necessary speech coding. An integrated user interface guides a user unfamiliar with the details of speech recognition or neural networks to quickly develop and test a neural network for phoneme recognition. To train the neural network, digitized voice data containing known phonemes that the user wants the neural network to ultimately recognize are processed by the integrated user interface. The digitized speech is segmented into phonemes with each segment being labelled with a corresponding phoneme code. Based on a user selected transformation method and transformation parameters, each segment is transformed into a series of multiple dimension vectors representative of the speech characteristics of that segment.

Abstract: An automated speech recognition system converts a speech signal into a compact, coded representation that correlates to a speech phoneme set. A number of different neural network pattern matching schemes may be used to perform the necessary speech coding. An integrated user interface guides a user unfamiliar with the details of speech recognition or neural networks to quickly develop and test a neural network for phoneme recognition. To train the neural network, digitized voice data containing known phonemes that the user wants the neural network to ultimately recognize are processed by the integrated user interface. The digitized speech is segmented into phonemes with each segment being labelled with a corresponding phoneme code. Based on a user selected transformation method and transformation parameters, each segment is transformed into a series of multiple dimension vectors representative of the speech characteristics of that segment.

Abstract: The invention relates to a voice recognition system which is robust to echoes and other background noise. An additional input signal describing a disturbance is evaluated such that during this additional recognition there is maximum suppression of information contained in the input signal. For this purpose, comparison vectors are formed which are continuously adapted to the instantaneous interference. The voice recognition system receives speech signals superimposed by noise signals. Additionally, a first spectral analysis unit produces first spectral values combined to first spectral vectors derived from disturbed speech signals. Estimates of the noise signals are also produced. A second spectral analysis unit produces second spectral values combined to second spectral vectors from the noise signal estimates.

Abstract: A transaction system has machine recognition of speech. It has dialogue control fed by the recognition, and speech generation fed by the dialogue control for outputting question and verifier statements from a repertoire set. A human-machine dialogue is executed until the dialogue control has recognized a viable transaction formulation with a plurality of user-provided slot fillers to specify the transaction. Dialogue control builds a directed and loopless status graph with nodes that each have their own slot filler and associated metric, and are interrelated through logic relations. The building can amend a node's metric and under control of conflict detection or lowering of a particular node's metric, discard the node in question and its filler, including of derived nodes and also of one-to-one derival nodes of the discarded node.

Abstract: A speech synthesizing apparatus for varying a speech characteristic condition is adapted to accept a speech request that does not have a speech characteristic condition and to synthesize a speech responsive thereto. A controlling portion accepts a plurality of speech requests; a speech synthesizing portion switches a plurality of speech characteristics for speech synthesis; a speaker outputs a speech corresponding to an output signal of the speech synthesizing portion; and a synthesizer characteristic table stores speech characteristic conditions synthesized by the speech synthesizing portion. The controlling portion can accept a speech request that does not have a speech characteristic condition. Then, the controlling portion selects an available speech characteristic condition from a synthesizer characteristic table and sends the selected speech characteristic condition to the speech synthesizer.

Abstract: A speech recognizing apparatus compares a speech command from a user with one of registration patterns stored in a storage unit in turn. Then if the speech command coincides with one of the registration patterns, the speech recognizing apparatus controls a predetermined electronic apparatus associated with an operation related to the registration pattern. If the speech command does not coincide with any one of the registration patterns, the speech recognizing apparatus stores into a memory the speech command as a new registration pattern in which the speech command is related to a manipulation of the electronic apparatus produced by the user immediately after speech command is produced.

Abstract: A recognition system includes a speech recognition processing unit for processing input speech signals to indicate similarity to predetermined patterns to be recognized. The recognition processing unit is arranged to repeatedly partition the input speech signal into a pattern-containing portion and, preceding and following the pattern-containing portions, noise or silence portions, and to identify a pattern corresponding to the pattern containing portion. An output supplies a recognition signal indicating recognition of one of the patterns. A pause detector detects the noise or silence portion which follows the pattern-containing portion. In response to its detection, a signal identifying the pattern currently corresponding to the pattern portion is supplied to the output. Also provided are similarly operating rejection portions.

Abstract: Fixed-format and coded control informations are received for generating announcements. The coded control informations select synthetic speech information items from a store. A speech generator under control of the control items forms a composite speech message. For a message containing both fixed items and variable items, fixed items are encoded in enriched phoneme notation. Variable items are encoded in straight phoneme notation. Items are provided in multiple versions. Each version has a respective different prosody pattern of pitch and/or rhythm of its phoneme sequence, and is selected by a multivalued context symbol adjoined or implicit to the associated control information element.

Abstract: A sound recognizer uses a feature value normalization process to substantially increase the accuracy of recognizing acoustic signals in noise. The sound recognizer includes a feature vector device which determines a number of feature values for a number of analysis frames, a min/max device which determines a minimum and maximum feature value for each of a number of frequency bands, a normalizer which normalizes each of the feature values with the minimum and maximum feature values resulting in normalized feature vectors, and a comparator which compares the normalized feature vectors with template feature vectors to identify one of the template feature vectors that most resembles the normalized feature vectors.

Abstract: An input speech signal is compressed and encoded by a speech encoding unit 3 and sent to an RF transmission processing unit 4 where it is channel-encoded, modulated and transmission-processed so as to be transmitted over an antenna 11. A signal received over the antenna 11 is reception-processed, demodulated and channel-decoded so as to be expanded and decoded a speech decoding unit 6. A recording/playback control unit 8 controls writing of a signal from the speech encoding unit 3 to a semiconductor memory 7 or readout of a signal from the semiconductor memory 7 to the speech encoding unit 3. This enables the semiconductor memory 7 to be efficiently utilized in single/dual communication without increasing the circuit construction.

Abstract: An audio/video (AV) information transaction system includes AV appliances installed in individual rooms, controllers associated with the AV appliances and a host computer which is linked with the controllers through a local area network (LAN), thereby forming a computer network. The host computer transfers control programs to the controllers. A controller in a room which runs the received control program sends a command signal of the demand of AV output reception to other controller in other room over the LAN cable, and the receiving controller operates the associated AV appliance and delivers the output AV signal to the demanding controller over the LAN cable.

Abstract: An efficient method for compressing audio and other sampled data signals without loss, or with a controlled amount of loss, is described. The compression apparatus contains a subset selector, an approximator, an adder, two derivative encoders, a header encoder, and a compressed block formatter. The decompression apparatus contains a compressed block parser, a header decoder, two integration decoders, an approximator, and an adder. The compressor first divides each block of input samples into a first subset and a second subset. The approximator uses the first subset samples to approximate the second subset samples. An error signal is created by subtracting the approximated second subset samples from the actual second subset samples. The first subset samples and error signal are separately encoded by the derivative encoders, which select the signal's derivative that requires the least amount of storage for a block floating point representation.

Abstract: Methods and apparatus for performing large-vocabulary speech recognition employing an integrated syntactic and semantic statistical language model. In an exemplary embodiment, a stochastic language model is developed using a hybrid paradigm in which latent semantic analysis is combined with, and subordinated to, a conventional n-gram paradigm. The hybrid paradigm provides an estimate of the likelihood that a particular word, chosen from an underlying vocabulary will occur given a prevailing contextual history. The estimate is computed as a conditional probability that a word will occur given an "integrated" history combining an n-word, syntactic-type history with a semantic-type history based on a much larger contextual framework. Thus, the exemplary embodiment seamlessly blends local language structures with global usage patterns to provide, in a single language model, the proficiency of a short-horizon, syntactic model with the large-span effectiveness of semantic analysis.

Abstract: A method and apparatus for the decoding of digital audio data encoded in accordance with layer 1 or 2 of the MPEG format. The inverse quantization of the quantizised data samples and the resealing of the inverse quantizised data samples takes place contemporaneously with windowing of the data samples transformed into the time domain and not contemporaneously with transformation of data samples from the frequency domain into the time domain using a matrix operation. The apparatus has fixed-wire frame unpacking and filter bank units. The frame unpacking unit is used for frame synchronization of a data stream, header decoding, reading of page information, inverse quantization of quantizised subband data samples and resealing of the inverse quantizised data samples. The filter bank unit is used for transformation of rescaled data samples from the frequency domain into the time domain using a matrix operation and for windowing of the data transformed into the time domain.