Abstract: Multiple microphone noise suppression apparatus and methods are described herein. The apparatus and methods implement a variety of noise suppression techniques and apparatus that can be selectively applied to signals received using multiple microphones. The microphone signals received at each of the multiple microphones can be independently processed to cancel echo signal components that can be generated from a local audio source. The echo cancelled signals may be processed by some or all modules within a signal separator that operates to separate or otherwise isolate a speech signal from noise signals. The signal separator can include a pre-processing de-correlator followed by a blind source separator. The output of the blind source separator can be post filtered to provide post separation de-correlation. The separated speech and noise signals can be non-linearly processed for further noise reduction, and additional post processing can be implemented following the non-linear processing.

Abstract: Enhancing noisy speech recognition accuracy by receiving geotagged audio signals that correspond to environmental audio recorded by multiple mobile devices in multiple geographic locations, receiving an audio signal that corresponds to an utterance recorded by a particular mobile device, determining a particular geographic location associated with the particular mobile device, selecting a subset of geotagged audio signals and weighting each geotagged audio signal of the subset based on whether the respective audio signal was manually uploaded or automatically updated, generating a noise model for the particular geographic location using the subset of weighted geotagged audio signals, where noise compensation is performed on the audio signal that corresponds to the utterance using the noise model that has been generated for the particular geographic location.

Abstract: A system and method is provided for sending a message type identifier through a speech codec (in-band) such as found in a wireless communication network. A first predetermined sequence with noise-like characteristics identifies a first message type. A second predetermined sequence with noise-like characteristics identifies a second message type.

Abstract: A method of pre-processing an audio signal transmitted to a user terminal via a communication network and an apparatus using the method are provided. The method of pre-processing the audio signal may prevent deterioration of a sound quality of the audio signal transmitted to the user terminal by pre-processing the audio signal, and by enabling a codec module, encoding the audio signal, to determine the audio signal as a speech signal. The method of pre-processing may include encoding the audio signal using a speech codec and decoding the encoded audio signal using the speech codec. A codec module, transmitting the decoded audio signal to the user terminal via the communication network, may determine whether a speech interval or a speechless interval with respect to at least one frame is included in the audio signal and transmit at least one parameter with respect to the at least one frame as a result of the determination.

Abstract: Techniques for introducing information into a data stream first obtains the spectral values of the short-term spectrum of the audio signal. Separately, information to be introduced are combined with a spread sequence obtaining a spread information signal, whereupon a spectral representation of the spread information is generated, then weighted with an established psychoacoustic maskable noise energy to generate a weighted information signal, wherein energy of the introduced information is substantially equal to or below the psychoacoustic masking threshold. The weighted information signal and the spectral values of the short-term spectrum of the audio signal are then summed and afterwards processed again to obtain a processed data stream including audio information and information to be introduced.

Abstract: A scalable coding apparatus is provided to suppress deterioration of a quality of a coded signal in a normal frame next to a frame compensated for the occurrence of a data loss. The scalable coding apparatus is provided with a core-layer coding section (11) to carry out core-layer coding for the n-th frame input audio signal, an ordinary coding section (121) to generate expanding-layer ordinary-coding layer L2(n) by carrying out ordinary-coding of an expanding layer for the input audio signal, a deterioration-compensation coding section (123) to generate an expanding-layer-deterioration coding data L2?(n) by carrying out compensation for quality deterioration of coded audio in a current frame due to a past frame loss, a judging section (125) to determine whether either the expanding-layer ordinary-coding data L2(n) or the expanding-layer deterioration-coding data L2?(n) should be output from the expanding-layer coding section (12) as expanding-layer coding data of the current frame.

Abstract: A scalable encoding apparatus capable of suppressing the quality degradation of a decoded signal without increasing the bit rate. In this apparatus, a core layer encoding part (101) and an extended layer encoding part (102) encode an input signal for each of audio frames. When a replacement determining part (103) determines that a degree to which the input signal changes between a preceding frame and a current frame is equal to or greater than a predetermined value or that a degree, to which the quality of the decoded signal is improved by an extended layer encoding process in the preceding frame, is equal to less than a predetermined level, a replacing part (105) replaces a part of an extended layer encoded data of the preceding frame by a core layer encoded data of the current frame. That is, a transmitting part (108) transmits, as a backup, the core layer encoded data of the current frame to a decoding end in advance.

Abstract: For an audio coding, noise suppression is applied to an original audio signal to obtain an audio signal with reduced noise. A coding mode is selected based on the audio signal with reduced noise. The original audio signal is then encoded using this selected coding mode.

Abstract: The present invention provides a method, a computer-software-product and an apparatus for enabling a determination of speech related audio data within a record of digital audio data. The method comprises steps for extracting audio features from the record of digital audio data, for classifying one or more subsections of the record of digital audio data, and for marking at least a part of the record of digital audio data classified as speech. The classification of the digital audio data record is performed on the basis of the extracted audio features and with respect to at least one predetermined audio class.

Abstract: A method of processing a decoded speech (DS) signal including successive DS frames, each DS frame including DS samples. The method comprises: adaptively filtering the DS signal to produce a filtered signal; gain-scaling the filtered signal with an adaptive gain updated once a DS frame, thereby producing a gain-scaled signal; and performing a smoothing operation to smooth possible waveform discontinuities in the gain-scaled signal.

Abstract: Encoding audio signals for Discontinuous with selecting an encoding mode for encoding the signal categorizing the signal into active segments having voice activity and non-active segments having substantially no voice activity by using categorization parameters depending on the quality of the selected encoding mode and encoding at least the active segments using the selected encoding mode that for a low quality encoding produce a lower number of “active” temporal section detections than for a high quality encoding mode, with comfort noise parameters producing less contrast from background noise for low quality encoding than for high quality modes.

Abstract: A speech recognition device and method configured to include a computer, for recognizing speech, including: a storage location for storing a feature quantity acquired from a speech signal for each frame; storage portions for storing acoustic model data and language model data; a echo speech component for generating echo speech model data from a speech signal acquired prior to a speech signal to be processed at the current time point and using the echo speech model data to generate adapted acoustic model data; and a processing component for utilizing the feature quantity, the adapted acoustic model data, and the language model data to provide a speech recognition result of the speech signal.

Abstract: A scalable encoding apparatus capable of reducing the bit rates of encoded parameters and also capable of efficiently encoding audio signals in which a plurality of harmonic structures are coexistent. In the apparatus, an MDCT analyzer MDCT analyzes an audio signal for converting/encoding processes. A pitch frequency converter determines an inverse of a pitch period to calculate a pitch frequency. A selector selects spectra located at frequencies that are integral multiples of the pitch frequency, and a second layer encoder encodes the selected spectra.

Abstract: A speech enhancement system improves the speech quality and intelligibility of a speech signal. The system includes a time-to-frequency converter that converts segments of a speech signal into frequency bands. A signal detector measures the signal power of the frequency bands of each speech segment. A background noise estimator measures a background noise detected in the speech signal. A dynamic noise reduction controller dynamically models the background noise in the speech signal. The speech enhancement renders a speech signal perceptually pleasing to a listener by dynamically attenuating a portion of the noise that occurs in a portion of the spectrum of the speech signal.

Abstract: An audio decoding system performs frame loss concealment (FLC) when portions of a bit stream representing an audio signal are lost within the context of a digital communication system. The audio decoding system employs two different FLC methods: one designed to perform well for music, and the other designed to perform well for speech. When a frame is deemed lost, the audio decoding system analyzes a previously-decoded audio signal corresponding to previously-decoded frames of an audio bit-stream. Based on the results of the analysis, the lost frame is classified as either speech or music. Using this classification, other signal analysis, and knowledge of the employed FLC methods, the audio decoding system selects the appropriate FLC method which then performs FLC on the lost frame.

Abstract: A scalable encoder enabling improvement of the encoding efficiency in the second layer and improvement of the quality of the original signal decoded using the encoding signal in the second layer. A predictive coefficient encoder of the scalable encoder has a predictive coefficient codebook where candidates of the predictive coefficient are recorded. After searching the predictive coefficient codebook, the scale factor of the first layer decoded signal inputted from a scale factor calculator is multiplied, and a predictive coefficient which most approximates the multiplication result to the scale factor of the original signal inputted from the scale factor calculator is determined and encoded, and the coded code is inputted to a multiplexer.

Abstract: A method of reducing noise in a speech signal involves converting the speech signal to the frequency domain using a fast fourier transform (FFT), creating a subset of selected spectral subbands, determining the appropriate gain for each subband, and interpolating the gains to match the number of FFT points. The converted speech signal is then filtered using the interpolated gains as filter coefficients, and an inverse FFT performed on the processed signal to recover the time domain output signal.

Abstract: A system and method for using bi-directional conversation data to improve signal presence detection are disclosed. The detector module is adapted to communicate with a signal enhancement module. The detector module collects data from a transmit direction of the connection and a receive direction of a data connection. The collected data from the transmit and the receive direction is used to classify at least one of data in the transmit direction and data in the receive direction. Responsive to the classification, the signal enhancement module enhances data in one of the transmit direction and the receive direction. Hence, data classification accuracy is improved by using data from both the transmit and receive directions. In one embodiment, the detector module applies a voice activity detection module (VAD) process to detect the presence or absence of voice data in the collected data.

Abstract: A headset is constructed to generate an acoustically distinct speech signal in a noisy acoustic environment. The headset positions a pair of spaced-apart microphones near a user's mouth. The microphones each receive the user's speech, and also receive acoustic environmental noise. The microphone signals, which have both a noise and information component, are received into a separation process. The separation process generates a speech signal that has a substantial reduced noise component. The speech signal is then processed for transmission. In one example, the transmission process includes sending the speech signal to a local control module using a Bluetooth radio.

Abstract: A method of and an apparatus for encoding/decoding an MPEG-4 bit sliced arithmetic coding (BSAC) audio bitstream having ancillary information. A time domain audio signal is converted to a frequency domain audio signal and quantized. A number of data bits is counted and a number of available bits per layer is obtained. The number of available bits per layer is modified considering the size of ancillary information. Actual audio data is encoded in units of layers and ancillary information is embedded in the encoded bitstream. A header is decoded and a layer structure of an audio bitstream is calculated to determine the size of the ancillary information as a difference between a size of data up to a top layer and a size of a frame. The ancillary information is extracted to improve meta data and sound quality of audio contents.

Abstract: In a communications system that demultiplexes user data words into multiple sub-words for encoding and decoding within different subword-processing paths, the minimum distance between bit errors in an extrinsic codeword can be increased by having corresponding interleavers/deinterleavers in the different subword-processing paths use different interleaving/deinterleaving algorithms.

Abstract: The invention relates to an audio encoder and decoder and methods for audio encoding and decoding. In a preferred encoder embodiment an audio signal is encoded by deterministic encoder means to form a first encoded signal part. A spectrum of the audio signal is determined and represented by an excitation pattern, i.e. spectral values corresponding to human auditory filters, as a second encoded signal part. A masking curve is also extracted based on the excitation pattern, thus improving encoding efficiency in terms of bit rate. In a preferred decoder the first encoded signal part is decoded by deterministic decoder means. A noise generator uses the decoded first signal part together with the second signal part, i.e. the excitation pattern for the original audio signal, to generate a noise signal. The noise signal is then added to the first decoded signal part to form an output audio signal. At the decoder side the masking curve is also extracted based on the second encoded signal part, i.e.

Abstract: A method for detecting voice activity comprises pre-processing a first frame in an audio frame sequence, receiving a subsequent frame as a current frame, calculating weighted linear prediction energy of the current frame based on Nth-order linear prediction coefficients, determining whether the current frame contains a noise or speech, if a speech is indicated, performing linear prediction analysis on the current frame to derive new Nth-order linear prediction coefficients and updating the coefficients with the derived one; if a nose is indicated and not the last frame, repeating the calculating and determining process. The corresponding device comprises a component for storing Nth-order linear prediction coefficients, a component for performing linear prediction analysis, a component for computing weighted linear prediction energy and a component for determining whether the current frame contains speech or noise based on calculated weighted linear prediction energy.

Abstract: A communication device includes memory, an input interface, a processing module, and a transmitter. The processing module receives a digital signal from the input interface, wherein the digital signal includes a desired digital signal component and an undesired digital signal component. The processing module identifies one of a plurality of codebooks based on the undesired digital signal component. The processing module then identifies a codebook entry from the one of the plurality of codebooks based on the desired digital signal component to produce a selected codebook entry. The processing module then generates a coded signal based on the selected codebook entry, wherein the coded signal includes a substantially unattenuated representation of the desired digital signal component and an attenuated representation of the undesired digital signal component. The transmitter converts the coded signal into an outbound signal in accordance with a signaling protocol and transmits it.

Abstract: To reduce noise in an input signal that may contain speech, first an estimate of the noise level in the signal is obtained. The level of the input signal is then compared with the noise level estimate signal to determine whether speech is dominant. Less aggressive noise reduction is applied to the input signal when speech is dominant than when only noise is present.

Abstract: An audio encoder, an audio decoder or an audio processor includes a filter for generating a filtered audio signal, the filter having a variable warping characteristic, the characteristic being controllable in response to a time-varying control signal, the control signal indicating a small or no warping characteristic or a comparatively high warping characteristic. Furthermore, a controller is connected for providing the time-varying control signal, which depends on the audio signal. The filtered audio signal can be introduced to an encoding processor having different encoding algorithms, one of which is a coding algorithm adapted to a specific signal pattern. Alternatively, the filter is a post-filter receiving a decoded audio signal.

Abstract: A system and method is provided for sending a message type identifier through a speech codec (in-band) such as found in a wireless communication network. A first predetermined sequence with noise-like characteristics identifies a first message type. A second predetermined sequence with noise-like characteristics identifies a second message type.

Abstract: An apparatus and method of low bit rate encoding and reproducing. The method includes transforming input audio signals in a time domain into spectral signals in a frequency domain, extracting important-spectrum components from the spectral signals in the frequency domain, and quantizing the important-spectrum components, extracting residual-spectrum components other than the important-spectrum components from the spectral signals in the frequency domain, and calculating and quantizing a noise level of the residual-spectrum components, and encoding the quantized important-spectrum components and the quantized noise level losslessly, and outputting encoded bitstreams.

Abstract: An input data is divided into blocks of predetermined time units and then further divided into a plurality of bands. Each band is coded and compressed to prepare a basic sub-block essential for data reproduction and a plurality of extension sub-blocks contributing to improvement of the quality of the reproduced data. A transmission timing during streaming delivery, remaining extension sub-blocks are delivered if the time period is still within a time period for delivering the same data block, but the remaining extension sub-blocks are not delivered and delivery of the basic sub-block of the subsequent data block begins if the delivery timing for the subsequent data block has arrived. In this way, a stable streaming delivery is achieved by a scalable data compression.

Abstract: A system and method for improving voice recognition processing at a server system that receives voice input from a remotely located user system. The user system includes a microphone, a processor that performs front-end voice recognition processing of the received user voice input, and a communication component configured to send the front-end processed user voice input to a destination wirelessly over a network. The server system includes a communication component configured to receive the sent front-end processed user voice input, and a processor configured to complete voice recognition processing of the sent front-end processed user voice input.

Abstract: An audio input signal is filtered using an adaptive filter to generate a prediction output signal with reduced noise, wherein the filter is implemented using a plurality of coefficients to generate a plurality of prediction errors and to generate an error from the plurality of prediction errors, wherein the absolute values of the coefficients are continuously reduced by a plurality of reduction parameters.

Abstract: A technique for computing perceptual noise in an audio signal that is computationally efficient. In one example embodiment, the technique includes computing perceptual noise in an input audio signal. The steps involve pre-computing NER (noise-to-excitation ratio) values associated with critical bands within a frame by zeroing out associated spectral coefficient values before the quantization loop, and also assuming bands with lower spectral energy than the band under consideration are zeroed out during quantization. When a critical band is zeroed out during quantization, the associated NER values which have been pre-computed are used in computing an overall perceptual distortion of the frame.

Abstract: Presented herein are systems and methods for generating an adaptive noise codebook for use with electronic speech systems. The noise codebook includes a plurality of entries which may be updated based on environmental noise sounds. The speech system includes a speech codebook and the adaptive noise codebook. The system identifies speech sounds in an audio signal using the speech and noise codebooks.

Abstract: Channel normalization for automatic speech recognition is provided. Statistics are measured from an initial portion of a speech utterance. Feature normalization parameters are estimated based on the measured statistics and a statistically derived mapping relating measured statistics and feature normalization parameters. In some examples, the measured statistics comprise measures of an energy from the initial portion of the speech utterance. In some examples, measures of the energy comprise extreme values of the energy.

Abstract: A noise reduction device including: an input signal spectrum obtaining unit that obtains an input signal spectrum by a subband unit based on a current frame of an input signal; an averaged spectrum obtaining unit that obtains an averaged spectrum of the input signal by averaging the input signal spectrum; an estimated noise spectrum obtaining unit that obtains an estimated noise spectrum estimated based on a past frame of the input signal by the subband unit; and an SN ratio obtaining unit that obtains an SN ratio by the subband unit, based on the averaged spectrum of the input signal obtained by the averaged spectrum obtaining unit, the estimated noise spectrum obtained by the estimated noise spectrum obtaining unit, and a function of the averaged spectrum of the input signal obtained by the averaged spectrum obtaining unit and the estimated noise spectrum obtained by the estimated noise spectrum obtaining unit.

Abstract: An “Interference Canceller” provides a computationally efficient real-time technique for removing stationary-tone interference from signals. Typical sources of stationary tone contamination of signals include noise from power wiring (i.e., 50/60 Hz or 400 Hz and their harmonics), frame or line frequencies from electronic devices, and noise from computer fans, hard disk drives, etc. In general, the Interference Canceller adaptively builds and updates a model of stationary tone interference in consecutive frames of an input signal. This adaptively updated model is then used to extrapolate and subtract noise from subsequent frames of the input signal to generate a “clean” output signal. This output signal exhibits significant attenuation of stationary tone interference without eliminating important portions of the underlying signal or distorting the underlying signal with artifacts such as musical noise or nonlinear distortions.

Abstract: A scalable coding apparatus is provided to suppress deterioration of a quality of a coded signal in a normal frame next to a frame compensated for the occurrence of a data loss. The scalable coding apparatus is provided with a core-layer coding section (11) to carry out core-layer coding for the n-th frame input audio signal, an ordinary coding section (121) to generate expanding-layer ordinary-coding layer L2(n) by carrying out ordinary-coding of an expanding layer for the input audio signal, a deterioration-compensation coding section (123) to generate an expanding-layer-deterioration coding data L2?(n) by carrying out compensation for quality deterioration of coded audio in a current frame due to a past frame loss, a judging section (125) to determine whether either the expanding-layer ordinary-coding data L2(n) or the expanding-layer deterioration-coding data L2?(n) should be output from the expanding-layer coding section (12) as expanding-layer coding data of the current frame.

Abstract: A system and process for muting the audio transmission from a location of a participant engaged in a multi-party, computer network-based teleconference when that participant is working on a keyboard, is presented. The audio is muted as it is assumed the participant is doing something other than actively participation in the meeting when typing on the keyboard. If left un-muted the sound of typing would distract the other participant in the teleconference.

Abstract: A method and apparatus identify a clean speech signal from a noisy speech signal. To do this, a clean speech value and a noise value are estimated from the noisy speech signal. The clean speech value and the noise value are then used to define a gain on a filter. The noisy speech signal is applied to the filter to produce the clean speech signal. Under some embodiments, the noise value and the clean speech value are used in both the numerator and the denominator of the filter gain, with the numerator being guaranteed to be positive.

Abstract: Generally described, the present invention is directed toward generating, maintaining, updating, and applying digital voice profiles. Voice profiles may be generated for individuals. The voice profiles include information that is unique to each individual and which may be applied to digital representations of that individual's voice to improve the quality of a transmitted digital representation of that individual's voice. A voice profile may include, but is not limited to, basic information about the individual, and filter definitions relating to the individuals voice patters, such as a frequency range and amplitude range. The voice profile may also include a speech definition that includes digital representations of the individual's unique speech patterns.

Abstract: In a noise estimation apparatus, a microphone converts sound into an electric signal and outputs the electric signal as a sound signal. A noise estimator performs estimation for estimating a magnitude of a noise component contained in the sound signal so as to generate an estimated noise signal. The noise estimator limits a minimum value of a noise level of the noise component contained in the sound signal during the estimation to a predetermined default value, and integrates the noise level having the limited minimum value to generate the estimated noise signal.

Abstract: A sound processing system including a user headset for use in tactical military operations provides integrated sound and speech analysis including sound filtering and amplification, sound analysis and speech recognition for analyzing speech and non-speech sounds and taking programmed actions based on the analysis, recognizing language of speech for purposes of one-way and two-way voice translation, word spotting to detect and identify elements of conversation, and non-speech recognition and identification. The headset includes housings with form factors for insulating a user's ear from direct exposure to ambient sounds with at least one microphone for receiving sound around the user, and a microphone for receiving user speech. The user headset can further include interconnections for connecting the headset with out systems outside of the headset, including target designation systems, communication networks, and radio transmitters.

Abstract: Methods and apparatus are presented for reducing the number of bits needed to represent an excitation waveform. An acoustic signal in an analysis frame is analyzed to determine whether it is a band-limited signal. A sub-sampled sparse codebook is used to generate the excitation waveform if the acoustic signal is a band-limited signal. The sub-sampled sparse codebook is generated by decimating permissible pulse locations from the codebook track in accordance with the frequency characteristic of the acoustic signal.

Abstract: The system for transmitting and receiving a wideband speech signal includes an A/D converter for receiving an analog speech signal to convert it into a digital speech signal, a transmitter analysis filter for receiving the digital speech signal and dividing it into a baseband signal and an enhancement residual band signal, a standard baseband encoder for accepting the baseband signal and coding it using an ITU-T encoder, an additional baseband encoder for reducing standard coding distortion in the baseband signal, an enhancement residual band encoder for coding a signal obtained by removing the coded baseband signal from the original digital speech signal, and an IP network interface for multiplexing the coded standard and additional baseband signals and enhancement residual band signal.

Abstract: For the transmission of a secondary information flow between a transmitter and a receiver, the secondary information flow is inserted at a parametric vocoder of the transmitter which generates a main information flow. The main information flow is a speech data flow encoding a speech signal and is transmitted from the transmitter to the receiver. Bits from the secondary information flow are inserted into only some of the frames of the main information flow, these frames being selected by a frame mask which is known to the transmitter and the receiver, and/or into a determined frame of the main information flow, by imposing a constraint on only some of the bits of the frame, these bits being selected by a bit mask known to the emitter and the receiver.

Abstract: A method and apparatus to reduce background noise in speech signals in order to improve the quality and intelligibility of processed speech. In mobile communications environment, speech signals are degraded by additive random noise. A randomness of the noise, which is often described in terms of its first and second order statistics, make it difficult to remove much of the noise without introducing background artifacts. This is particularly true for lower signal to background noise ratios. The method and apparatus provides noise reduction without any knowledge of the signal to background noise ratio.

Abstract: This invention relates to a computer-based method and system for facilitating the retrieval, classification, and distribution of information. In one embodiment, a method for providing information comprises providing a plurality of entities, each having an entity type, providing a plurality of relationships among the entities, each relationship having a relationship type and direction, and constructing an entity-relationship network comprising the entities and relationships. The method further includes receiving a plurality of information items, facilitating the association of the information items with at least one corresponding entity, receiving a request for information items associated with a selected one of the entities, determining a subset of the entities based on the selected entity, the relationships, the relationship types and the relationship directions, and providing the information items associated with the subset of the entities.

Abstract: A multimedia device can be used to play audio. Speech in an environment proximate to a multimedia device can be detected. The detected speech can be recorded. The playing of the audio can be paused. The recorded speech can be audibly presented. A condition to resume the paused audio can be detected. The paused audio can be resumed from the previously paused position.

Abstract: A speech recognition device and method configured to include a computer, for recognizing speech, including: a storage location for storing a feature quantity acquired from a speech signal for each frame; storage portions for storing acoustic model data and language model data; a echo speech component for generating echo speech model data from a speech signal acquired prior to a speech signal to be processed at the current time point and using the echo speech model data to generate adapted acoustic model data; and a processing component for utilizing the feature quantity, the adapted acoustic model data, and the language model data to provide a speech recognition result of the speech signal.

Abstract: A method and apparatus for detecting speech segments of a speech signal processing device is provided. A critical band is divided into a certain number of regions according to noise frequency characteristics, a signal threshold and a noise threshold are set for each of the regions, and it is determined whether each frame is a speech segment or noise segment by comparing the log energy calculated for each region to the corresponding signal threshold and noise threshold. Therefore, a speech segment can be detected rapidly and accurately by using a small number of operations even in a noise environment.