Abstract: A sound processing apparatus is provided for estimating the power of background noise using a directional sound receiving technology using a plurality of sound receiving units, computing a gain control value on the basis of the estimated power of background noise and a predetermined power target value, and outputting the gain control value, so that a delay time of starting gain control can be reduced, and a slow response of a speech recognition application program or degradation of the speech quality of a voice communication program can be prevented.

Abstract: A handsfree device, which is coupled to a data processing device, may be operable to monitor at least one audio stream for occurrence of at least one keyword. Upon recognition of the at least one keyword, the handsfree device may establish a first connection between the handsfree device and the data processing device for launching a voice interface in the data processing device. The handsfree device may send audio data received after the recognition of the at least one keyword to the data processing device, via the first connection for responding to the audio data via the voice interface. During a keyword configuration operation, the handsfree device may send at least one inputted keyword to the data processing device for recording. The handsfree device may receive, via a second connection, the recorded at least one keyword from the data processing device for keyword configuration of the handsfree device.

Abstract: The present invention relates to methods and devices for encoding and decoding digital audio signals, e.g. a speech signal. An audio coder and a decoder are provided wherein a modeller adds a first distribution model obtained from model parameters of past segments of the digital audio signal and a fixed distribution model, each of the models being multiplied by a weighting coefficient, for obtaining a combined distribution model. The weighting coefficients are selected to minimize a code length of a current segment of the digital audio signal. As the combined distribution model is a sum of several distribution models, wherein at least some of the models is based on the model parameters, flexibility is introduced in the signal model used to encode the digital audio signal. Thus, an audio coder and decoder providing a low bit rate in average, low bit rate variations and low error propagation are provided.

Abstract: A computerized system for advising one communicant in electronic communication between two or more communicants has apparatus monitoring and recording interaction between the communicants, software executing from a machine-readable medium and providing analytics, the software functions including rendering speech into text, and analyzing the rendered text for topics, performing communicant verification, and detecting changes in communicant emotion. Advice is offered to the one communicant during the interaction, based on results of the analytics.

Abstract: According to one aspect, a method for coreference resolution is provided. In one embodiment, the method includes receiving a segment of text that includes mentions corresponding to entities. A first feature vector is generated based on one or more features associated with a first mention, and a second feature vector is generated based on based on one or more features associated with a second mention. A measure of similarity between the first feature vector and second feature vector is computed and, based on the computed measure of similarity, it is determined if the first mention and the second mention both correspond to the same entity.

Abstract: A sub-band processing system that reduces computational complexity and memory requirements includes a processor and a local or distributed memory. Logic stored in the memory partitions a frequency spectrum of bins into a smaller number of sub-bands. The logic enables a lossy compression by designating a magnitude and a designated or derived phase of each bin in the frequency spectrum as representative. The logic renders a lossless compression by decompressing the lossy compressed data and providing lost data based on original spectral relationships contained within the frequency spectrum.

Abstract: An audio decoder for providing a decoded representation of an audio content on the basis of an encoded representation of the audio content comprises a linear-prediction-domain decoder core configured to provide a time-domain representation of an audio frame on the basis of a set of linear-prediction domain parameters associated with the audio frame and a frequency-domain decoder core configured to provide a time-domain representation of an audio frame on the basis of a set of frequency-domain parameters, taking into account a transform window out of a set comprising a plurality of different transform windows. The audio decoder comprises a signal combiner configured to overlap-and-add-time-domain representations of subsequent audio frames encoded in different domains, in order to smoothen a transition between the time-domain representations of the subsequent frames.

Abstract: A computerized readable apparatus for presentation of information, including contextually related secondary content. In one embodiment, the apparatus comprises a computer readable medium having at least one computer program disposed thereon, the at least one program being configured to facilitate ad hoc communication with a personal electronic device of a user, and provide the user with requested information (such as for example direction to a desired business or other entity). At least a portion of the information is obtained via a wireless link with a remote server.

Abstract: In a method and a system (20) for creating or updating entries in a speech recognition (SR) lexicon (7) of a speech recognition system, said entries mapping speech recognition (SR) phoneme sequences to words, said method comprising entering a respective word, and in the case that the word is a new word to be added to the SR lexicon, also entering at least one associated SR phoneme sequence through input means (26), it is provided that the SR phoneme sequence associated with the respective word is converted into speech by phoneme to speech conversion means (4.4), and the speech is played back by playback means (28), to control the match of the phoneme sequence and the word.

Abstract: Speech processing is disclosed for an apparatus having a main processing unit, a memory unit, and one or more co-processors. Memory maintenance and voice recognition result retrievals upon execution are performed with a first main processor thread. Voice detection and initial feature extraction on the raw data are performed with a first co-processor. A second co-processor thread receives feature data derived for one or more features extracted by the first co-processor thread and information for locating probability density functions needed for probability computation by a speech recognition model and computes a probability that the one or more features correspond to a known sub-unit of speech using the probability density functions and the feature data. At least a portion of a path probability that a sequence of sub-units of speech correspond to a known speech unit is computed with a third co-processor thread.

Abstract: Architecture for integrating application-dependent information into a constraints component at deployment time or when available. In terms of a general grammar, the constraints component can include or be a general grammar that comprises application-independent information and is structured in such a way that application-dependent information can be integrated into the general grammar without loss of fidelity. The general grammar includes a probability space and reserves a section of the probability space for the integration of application-dependent information. An integration component integrates the application-dependent information into the reserved section of the probability space for recognition processing. The application-dependent information is integrated into the reserved section of the probability space at deployment time or when available. The general grammar is structured to support the integration and improve the overall system.

Abstract: Provided is a multi-channel audio decoding method and apparatus therefor, the method involving decoding filter bank coefficients of a plurality of bands from a bitstream having a predetermined format; performing frequency transformation on the decoded filter bank coefficients of the plurality of bands, with respect to each of the plurality of bands; compensating for a phase of each of the plurality of bands according to a predetermined phase compensation value, and serially band-synthesizing the frequency-transformed coefficients of each of the plurality of phase-compensated bands on a frequency domain; and decoding a multi-channel audio signal from the band-synthesized frequency-transformed coefficients.

Abstract: A voice mixing device for mixing a plurality of voice signals, comprises: a speaker selection unit selecting at least one voice signal among said plurality of voice signals; a full signal adder unit adding all of at least one voice signal selected by said speaker selection unit; respective subtractor unit subtracting only one of said selected voice signals from an addition result of said full signal adder unit; a common noise suppression unit suppressing noise of a common voice signal, being an addition result of said full signal adder unit; individual noise suppression unit suppressing noise of respective individual voice signals, being subtraction results of said subtractor unit; and memory switching unit copying information of noise suppression obtained in said common noise suppression unit based on a selection result of said speaker selection unit, to information of noise suppression in said individual noise suppression unit.

Abstract: The equipment comprises at least one computer and a material features acquisition system operable to detect a plurality of material features. The features are then evaluated according to rules that capture the multidiscipline knowledge of experts and are already inputted into the computer. The computer iterations are processed until an acceptable conclusion is made regarding the condition of the material under evaluation.

Abstract: An apparatus for classifying an input audio signal into audio contents of a first and second class, comprising an audio segmentation module adapted to segment said input audio signal into segments of a predetermined length; a feature computation module adapted to calculate for the segments features characterizing said audio input signal; a threshold comparison module adapted to generate a feature vector for each of said one or more segments based on a plurality of predetermined thresholds, the thresholds including for each of the audio contents of the first class and of the second class a substantially near certainty threshold, a substantially high certainty threshold, and a substantially low certainty threshold; and a classification module adapted to analyze the feature vector and classify each one of said one or more segments as audio contents of the first class, of the second class, or as non-decisive audio contents.

Abstract: A method for performing packet loss or Frame Erasure Concealment (FEC) for a speech coder receives encoded frames of compressed speech information transmitted from an encoder. The method determines whether an encoded frame has been lost, corrupted in transmission, or erased, synthesizes properly received frames, and decides on an overlap-add window to use in combining a portion of the synthesized speech signal with a subsequent speech signal resulting from a received and decoded packet, where the size of the overlap-add window is based on the unavailability of packets. If it is determined that an encoded frame has been lost, corrupted in transmission, or erased, the method performed an overlap-add operation on the portion of the synthesized speech signal and the subsequent speech signal, using the decided-on overlap-add window.

Abstract: A method for estimating high-order Mel Frequency Cepstral Coefficients, the method comprising initializing any of N?L high-order coefficients (HOC) of an MFCC vector of length N having L low-order coefficients (LOC) to a predetermined value, thereby forming a candidate MFCC vector, synthesizing a speech signal frame from the candidate MFCC vector and a pitch value, and computing an N-dimensional MFCC vector from the synthesized frame, thereby producing an output MFCC vector.

Abstract: Some methods may involve receiving a frame of encoded audio data that includes transform coefficient data. The transform coefficient data may include exponent data and mantissa data. The mantissa data may include mantissa values that were encoded with uniform or non-uniform boundaries of quantization intervals. The mantissa values may be reconstructed based, at least in part, on exponent profile data. Based on the exponent profile data, statistics regarding the pre-quantization mantissas values may be inferred. The exponent profile data may include exponent differential data. Some such exponent differential data may be exponent difference pairs, though more than two exponent differential data points may be evaluated in alternative methods. At each frequency bin, mantissa value reconstruction may be conditioned on the exponent differential data, e.g., on the exponent difference pairs.

Abstract: A system and methods of subtraction of a shaped component of a noise reduction spectrum from a combined signal are disclosed. In an embodiment, a method includes identifying a selected frequency component using a corresponding frequency component of a noise sample spectrum. A noise set is comprised of the noise sample spectrum. The method further includes forming a shaped component of a noise reduction spectrum using a processor and a memory based on a combined signal spectrum and the selected frequency component. The method also includes subtracting the shaped component of the noise reduction spectrum from the combined signal spectrum.

Abstract: A speech data retrieval apparatus (10) includes a speech database (1), a speech recognition unit (2), a confusion network creation unit (3), an inverted index table creation unit (4), a query input unit (6), a query conversion unit (7) and a label string check unit (8). The speech recognition unit (2) reads speech data from the speech database (1), carries out a speech recognition process with respect to the read speech data, and outputs a result of speech recognition process as a lattice in which a phoneme, a syllable, or a word is a base unit. The confusion network creation unit (3) creates a confusion network based on the output lattice and outputs the result of speech recognition process as the confusion network. The inverted index table creation unit (4) creates an inverted index table based on the output confusion network.