Abstract: A technique for correcting the voice spectral deformations introduced by a communication network. Prior to the operation of equalization of the voice signal of a speaker, the constitution of classes of speakers is communicated, with one voice reference per class. Then, for a given speaker, the classification of this speaker is communicated, that is to say his allocation to a class from predefined classification criteria in order to make a voice reference which is closest to his own correspond to him. Then, for that given speaker, communicating the equalization of the digitized signal of the voice of the speaker carried out with, as a reference spectrum, the voice reference of the class to which the speaker has been allocated. This technique applies to the correction of the timbre of the voice in switched telephone networks, in ISDN networks and in mobile networks.

Abstract: In one embodiment, the present invention is a method and apparatus for active noise cancellation. In one embodiment, a method for recognizing user speech in an audio signal received by a media system (where the audio signal includes user speech and ambient audio output produced by the media system and/or other devices) includes canceling portions of the audio signal associated with the ambient audio output and applying speech recognition processing to an uncancelled remainder of the audio signal.

Abstract: A method and apparatus determine a channel response for an alternative sensor using an alternative sensor signal, an air conduction microphone signal. The channel response and a prior probability distribution for clean speech values are then used to estimate a clean speech value.

Abstract: At the coder side, bits of samples of each frame of an input digital signal are concatenated every digit common to the samples across each frame to generate equi-order bit sequences, which are output as packets. At the decoding side, the input equi-order sequences are arranged inversely to their arrangement at the coder side to reconstruct sample sequences. When a packet dropout occurs, a missing information compensating part 430 correct the reconstructed sample sequences in a manner to reduce an error between the spectral envelope of the reconstructed sample sequence concerned and a known spectral envelope.

Abstract: A packet loss concealment method and system is described that attempts to reduce or eliminate destructive interference that can occur when an extrapolated waveform representing a lost segment of a speech or audio signal is merged with a good segment after a packet loss. This is achieved by guiding a waveform extrapolation that is performed to replace the bad segment using a waveform available in the first good segment or segments after the packet loss. In another aspect of the invention, a selection is made between a packet loss concealment method that performs the aforementioned guided waveform extrapolation and one that does not. The selection may be made responsive to determining whether the first good segment or segments after the packet loss are available and also to whether a segment preceding the lost segment and the first good segment following the lost segment are deemed voiced.

Abstract: A technique is described herein for updating a state of a decoder configured to decode a series of frames representing an encoded audio signal. In accordance with the technique, an output audio signal associated with a lost frame in the series of frames is synthesized. The decoder state is set to align with the synthesized output audio signal at a frame boundary. An extrapolated signal is generated based on the synthesized output audio signal. A time lag is calculated between the extrapolated signal and a decoded audio signal associated with a first received frame after the lost frame in the series of frames, wherein the time lag represents a phase difference between the extrapolated signal and the decoded audio signal. The decoder state is then reset based on the time lag.

Abstract: A technique is described herein for reducing audible artifacts in an audio output signal generated by decoding a received frame in a series of frames representing an encoded audio signal in a predictive coding system. In accordance with the technique, it is determined if the received frame is one of a predefined number of received frames that follow a lost frame in the series of the frames. Responsive to determining that the received frame is one of the predefined number of received frames, at least one parameter or signal associated with the decoding of the received frame is altered from a state associated with normal decoding. The received frame is then decoded in accordance with the at least one parameter or signal to generate a decoded audio signal. The audio output signal is then generated based on the decoded audio signal.

Abstract: An apparatus for canceling an echo signal in a mobile terminal of a mobile communication system. A double talk detector (DTD) receives a first signal by canceling an estimated echo signal from a signal received through a microphone, outputs the first signal, and outputs the first and a second signal comprising a background noise signal and a residual echo signal during a non-double talk. An Auto-Regressive (AR) analysis and inverse filtering unit receives the second signal from the DTD, and whitens the second signal. A pitch analysis and inverse filtering unit receives the whitened signal, and cancels a pitch value remaining therein by performing pitch analysis and inverse filtering on the whitened signal. A noise canceller receives the pitch-cancelled whitened signal and the first signal output from the DTD, canceling a residual echo signal and a background noise signal from the first signal using the pitch-cancelled whitened signal.

Abstract: Methods and devices for dynamically adjusting a multi-band signal-modification profile based on a psychoacoustic model are disclosed. In one arrangement, the encoding parameter side information is used to estimate encoding noise of an encoded signal. The signal spectrum of the signal is estimated. Adjustments to the multi-band signal-modification profile are determined using the estimated noise and signal spectrum and a psychoacoustic profile.

Abstract: A method, apparatus and system that receives speech commands at a remote control device microphone, digitizes those input speech commands, compresses the digitized speech commands, multiplexes control commands with the compressed digitized speech commands, and transmits the compressed digitized speech commands to an electronic device, such as a digital home communication terminal (DCHT). The electronic device decompresses and interprets the speech commands to allow the remote control operator to control the electronic device. Because speech recognition is performed at the electronic device, rather than at the remote control device, the remote control does not have to interpret and transmit infrared signals that represent user commands. This simplifies the processing and voice recognition capabilities required by the remote control.

Abstract: A method for calculating the amplication factor, which co-determines the volume, for a speech signal transmitted in encoded form includes dividing the speech signal into short temporal signal segments. The individual signal segments are encoded and transmitted separately from each other, and the amplication factor for each signal segment is calculated, transmitted and used by the decoder to reconstruct the signal. The amplication factor is determined by minimizing the value E(g_opt2)=(1?a)*f1(g_opt2)+a*f2(g_opt2), the weighting factor a being determined taking into account both the periodicity and the stationarity of the encoded speech signal.

Abstract: A method and system are provided for processing an extrapolated signal including a number of consecutive replacement frames. The method comprises attenuating a portion of the extrapolated signal when the extrapolated signal reaches a predetermined duration. The attenuating produces an output signal having an attenuated portion, wherein the output signal includes the number of consecutive replacement frames. Each of the consecutive frames within the attenuated portion is attenuated by applying an attenuation window with a starting magnitude value of approximately 1 and including a unique ending magnitude. The unique ending magnitudes decrease over time.

Abstract: Circuitry and a method compensate the erasure of speech signal data or similar periodic signal data, by substitution using past periodic signal data input. After a predetermined number of latest periodic signal data have been saved, whether or not an erasure occurs is determined with every periodic signal data sequence, which is a unit of processing. When an erasure occurs, one of periodic signal data sequences saved, which lies in a determined segment to be used, is used to generate synthetic data for substitution. The position of the segment to be used is determined such that when the erasure continues over units of processing, the position sequentially varies gradually for each processing units.

Abstract: Provided are a speech restoration system and method for concealing packet losses.

Abstract: A system and method for voice quality analysis include the ability to receive packets in a voice stream and to generate a receipt indicator for the packets. The system and method also include the ability to substitute a reference voice sample for the voice data in the packets and to compare the voice data in the voice-substituted packets to the reference voice sample to determine voice quality.

Abstract: An enhancement system improves the estimate of noise from a received signal. The system includes a spectrum monitor that divides a portion of the signal at more than one frequency resolution. Adaptation logic derives a noise adaptation factor of the received signal. A plurality of devices tracks the characteristics of an estimated noise in the received signal and modifies multiple noise adaptation rates. Weighting logic applies the modified noise adaptation rates derived from the signal divided at a first frequency resolution to the signal divided at a second frequency resolution.

Abstract: A mobile device includes an air conduction microphone and an alternative sensor that provides an alternative sensor signal indicative of speech. A communication interface permits the mobile device to communicate directly with other mobile devices.

Abstract: A method and apparatus for noise suppression is described herein. The channel gain is controlled based on a degree of variability of the background noise. The noise variability estimate is used in conjunction with a variable attenuation concept to produce a family of gain curves that are adaptively suited for a variety of combinations of long-term peak SNR and noise variability. More specifically, a measure of the variability of the background noise is used to provide an optimized threshold that reduces the occurrence of non-stationary background noise entering into the transition region of the gain curve.

Abstract: A method for determining intensity characteristics of background noise during speech pauses of speech signals includes determining a proportion of speech pauses in the undisturbed source speech signal so as to define a frequency threshold. The disturbed speech signal is divided into short successive signal elements, an intensity value is determined for each of the signal elements, and a cumulative relative frequency distribution is formed from the determined intensity values of the signal elements. The cumulative relative frequency distribution is used to determine an intensity threshold value which corresponds to the defined frequency threshold. At least one intensity characteristic of the background noise during the speech pauses is determined using a region of the cumulative relative frequency distribution below the intensity threshold value.

Abstract: The present invention provides a method and apparatus for accelerated handwritten symbol recognition in a pen based tablet computer. In one embodiment, handwritten symbols are translated into machine readable characters using special purpose hardware. In one embodiment, the special purpose hardware is a recognition processing unit (RPU) which performs feature extraction and recognition. A user inputs the handwritten symbols and software recognition engine preprocesses the input to a reduced form. The data from the preprocessor is sent to the RPU which performs feature extraction and recognition. In one embodiment, the RPU has memory and the RPU operates on data in its memory. In one embodiment, the RPU uses a hidden Markov model (HMM) as a finite state machine that assigns probabilities to a symbol state based on the preprocessed data from the handwritten symbol. In another embodiment, the RPU recognizes collections of symbols, termed “wordlets,” in addition to individual symbols.

Abstract: A method and apparatus are provided for identifying a noise environment for a frame of an input signal based on at least one feature for that frame. To identify the noise environment, a probability for a noise environment is determined by applying the noisy input feature vector to a distribution of noisy training feature vectors. In one embodiment, each noisy training feature vector in the distribution is formed by modifying a set of clean training feature vectors. In one embodiment, the probabilities of the noise environments for past frames are included in the identification of an environment for a current frame. In one embodiment, a correction vector is then selected based on the identified noise environment.

Abstract: The present invention relates to a new method for enhancement of source coding systems using high-frequency reconstruction. The invention teaches that tonal signals can be classified as either pulse-train-like or non-pulse-train-like. Relying on this classification, significant improvements on the perceived audio quality can be obtained by adaptive switching of transposers. The invention shows that the so-switched transposers must have fundamental differences in their characteristics.

Abstract: A method and apparatus are provided for reducing noise in a training signal and/or test signal. The noise reduction technique uses a stereo signal formed of two channel signals, each channel containing the same pattern signal. One of the channel signals is “clean” and the other includes additive noise. Using feature vectors from these channel signals, a collection of noise correction and scaling vectors is determined. When a feature vector of a noisy pattern signal is later received, it is multiplied by the best scaling vector for that feature vector and the best correction vector is added to the product to produce a noise reduced feature vector. Under one embodiment, the best scaling and correction vectors are identified by choosing an optimal mixture component for the noisy feature vector. The optimal mixture component being selected based on a distribution of noisy channel feature vectors associated with each mixture component.

Abstract: The invention provides a method and system for dynamically estimating background noise. The system includes a portable communication device, a vocoder, and a voice activated detector. Based on information received by the portable communication device, the vocoder determines parameters related to incoming information including a voicing mode indicative of the periodicity of incoming information. The voice activated detector then compares the voicing mode to a threshold to determine whether a background noise estimate should be updated.

Abstract: A comfort noise generator (104) suitable for use in a communication system includes a finite impulse response (FIR) filter (136), a random number generator (140), and a coefficient updater (138). The coefficient updater (138) determines an updated set of filter coefficients (142) based on the signal frame of the input signal (102). The updated set of filter coefficients (142) is output to the FIR filter (136). The FIR filter (136) shapes a white noise signal (146) supplied by the random number generator (140) to provide a simulated background noise signal, or comfort noise signal (122). The comfort noise signal (122) is selectively output from an echo suppression system or corresponding method to overwrite or suppress reflected residual echoes.

Abstract: A method and apparatus for distributed speech recognition serve to mitigate the effect of transmission errors. The method comprises the steps of identifying speech recognition parameters which have been subjected to a transmission error, and processing data to be sent to the speech recognition decoder to ensure that any speech recognition parameters which have been subjected to such an error are excluded from back-end processing. The speech recognition parameters which have been subjected to transmission errors are excluded from back-end processing by replacing those parameters with data which is selected so as to be rejected by the speech recognition decoder as abnormal and/or non-speechlike.

Abstract: The invention concerns a method and apparatus for performing packet loss or Frame Erasure Concealment (FEC) for a speech coder that does not have a built-in or standard FEC process. A receiver with a decoder receives encoded frames of compressed speech information transmitted from an encoder. A lost frame detector at the receiver determines if an encoded frame has been lost or corrupted in transmission, or erased. If the encoded frame is not erased, the encoded frame is decoded by a decoder and a temporary memory is updated with the decoder's output. A predetermined delay period is applied and the audio frame is then output. If the lost frame detector determines that the encoded frame is erased, a FEC module applies a frame concealment process to the signal. The FEC processing produces natural sounding synthetic speech for the erased frames.

Abstract: A decoding method for coded data representing original data. Corrupted data is detected and replaced with buffered data. The buffered data is stored in the buffer a time interval corresponding to an estimated periodicity or an integer multiple thereof before the corrupted data was received. The estimated periodicity is determined by estimating the periodicity of the original data represented by the corrupted data.

Abstract: Acoustic signals are analyzed by two-dimensional (2-D) processing of the one-dimensional (1-D) speech signal in the time-frequency plane. The short-space 2-D Fourier transform of a frequency-related representation (e.g., spectrogram) of the signal is obtained. The 2-D transformation maps harmonically-related signal components to a concentrated entity in the new 2-D plane (compressed frequency-related representation). The series of operations to produce the compressed frequency-related representation is referred to as the “grating compression transform” (GCT), consistent with sine-wave grating patterns in the frequency-related representation reduced to smeared impulses. The GCT provides for speech pitch estimation. The operations may, for example, determine pitch estimates of voiced speech or provide noise filtering or speaker separation in a multiple speaker acoustic signal.

Abstract: A method and apparatus for reducing noise in a speech signal. A handset or remote unit provides to users with a hearing deficiency, a first mode of operation where noise suppressant/speech enhancement algorithms are used during any auditory-related service. There is also provided, in a related mode of operation, speech filtering for reducing noise in a speech signal received through the microphone and outputting the filtered sound to the speaker. The handset includes a microphone for receiving an auditory sound, a receiver for receiving an auditory signal and a speech filter for suppressing noise in the auditory signal and sound. The speech filter also may be configured to shift the frequency and/or alter the intensity of the auditory signal and sound. The speaker is used for amplifying and outputting the enhanced speech component as an audible sound.

Abstract: Recognizing a stream of speech received as speech vectors over a lossy communications link includes constructing for a speech recognizer a series of speech vectors from packets received over a lossy packetized transmission link, wherein some of the packets associated with each speech vector are lost or corrupted during transmission. Each constructed speech vector is multi-dimensional and includes associated features. Potentially corrupted features within the speech vector are indicated to the speech recognizer when present. Speech recognition is attempted at the speech recognizer on the speech vectors when corrupted features are present. This recognition may be based only on certain or valid features within each speech vector. Retransmission of a missing or corrupted packet is requested when corrupted values are indicated by the indicating step and when the attempted recognition step fails.

Abstract: A distributed speech recognition system includes a noise floor estimator to provide a noise floor estimate to a feature extractor which provides a parametric representation of the noise floor estimate. An encoder is included to to generate an encoded parametric representation of the noise floor estimate. A front-end controller is also included to determine when at least one of the noise floor estimator, the feature extractor, and the encoder is to be turned on or off and to determine when the noise floor estimator is to provide the noise floor estimate to the feature extractor. Additionally, a decoder is included to generate a decoded parametric representation of the noise floor estimate. A noise model generator creates a statistical model of noise feature vectors based on the decoded parametric representation of the noise floor estimate.

Abstract: A method and apparatus estimate additive noise in a noisy signal using incremental Bayes learning, where a time-varying noise prior distribution is assumed and hyperparameters (mean and variance) are updated recursively using an approximation for posterior computed at the preceding time step. The additive noise in time domain is represented in the log-spectrum or cepstrum domain before applying incremental Bayes learning. The results of both the mean and variance estimates for the noise for each of separate frames are used to perform speech feature enhancement in the same log-spectrum or cepstrum domain.

Abstract: A sound processing method and apparatus are provided, which are capable of performing sound processing on input audio signals containing a plurality of signal components being different in desired sound processing conditions, in a manner that allows natural sound to be reproduced. An input audio signal of at least one system is separated into a plurality of separated signal components, and each signal component of at least part of the plurality of separated signal components is subjected to individual sound processing according to the signal component, and the plurality of separated signal components are outputted as at least one audio signal after each signal component of the at least part thereof is subjected to the individual sound processing. The plurality of separated signal components are synthesized into a synthesized audio signal, which is then outputted, or alternatively, the plurality of separated signal components are outputted separately as audio signals.

Abstract: An estimate is made of the power of a speech portion of a speech signal that includes speech portions separated by non-speech portions, the power for the speech portion being estimated based on a power envelope that spans the speech portion. The gain of an automatic gain control is not adjusted during the speech portions.

Abstract: A method and system are provided for removing discontinuities associated with synthesizing a corrupted frame output from a decoder including one or more predictive filters. The corrupted frame is representative of one segment of a decoded signal. The method comprises copying a first number of stored samples of the decoded signal in accordance with a time lag and a scaling factor, and calculating a first number of ringing samples output from at least one of the filters.

Abstract: A method and apparatus estimate additive noise in a noisy signal using an iterative technique within a recursive framework. In particular, the noisy signal is divided into frames and the noise in each frame is determined based on the noise in another frame and the noise determined in a previous iteration for the current frame. In one particular embodiment, the noise found in a previous iteration for a frame is used to define an expansion point for a Taylor series approximation that is used to estimate the noise in the current frame. In one embodiment, noise estimation employs a recursive-Expectation-Maximization framework with a maximum likelihood (ML) criteria. In a further embodiment, noise estimation employs a recursive-Expectation-Maximization framework based on a MAP (maximum a posterior) criteria.

Abstract: There are provided a method and an apparatus for reducing random, continuous, non-stationary noise in audio signals, the noisy audio signal being filtered by means of a predetermined filter function. The filter function is determined dynamically having regard to the current properties of the noisy audio signal and/or its constituent parts, and the filter function is also limited dynamically having regard to the current properties of the noise component contained in the noisy audio signal.

Abstract: A method and apparatus are provided for reducing noise in a signal. Under one aspect of the invention, a correction vector is selected based on a noisy feature vector that represents a noisy signal. The selected correction vector incorporates dynamic aspects of pattern signals. The selected correction vector is then added to the noisy feature vector to produce a cleaned feature vector. In other aspects of the invention, a noise value is produced from an estimate of the noise in a noisy signal. The noise value is subtracted from a value representing a portion of the noisy signal to produce a noise-normalized value. The noise-normalized value is used to select a correction value that is added to the noise-normalized value to produce a cleaned noise-normalized value. The noise value is then added to the cleaned noise-normalized value to produce a cleaned value representing a portion of a cleaned signal.

Abstract: The invention concerns a method and apparatus for performing packet loss or Frame Erasure Concealment (FEC) for a speech coder that does not have a built-in or standard FEC process. A receiver with a decoder receives encoded frames of compressed speech information transmitted from an encoder. A lost frame detector at the receiver determines if an encoded frame has been lost or corrupted in transmission, or erased. If the encoded frame is not erased, the encoded frame is decoded by a decoder and a temporary memory is updated with the decoder's output. A predetermined delay period is applied and the audio frame is then output. If the lost frame detector determines that the encoded frame is erased, a FEC module applies a frame concealment process to the signal. The FEC processing produces natural sounding synthetic speech for the erased frames.

Abstract: Wideband speech signals must be converted to narrowband speech signals if the transmission medium or the destination terminal is constructed with narrowband constraints. A typical wideband-to-narrowband conversion method is the elimination of frequencies above 3400 Hz using a low pass filter and a down sampler. However, this method produces a muffled speech sound since the resulting narrowband signal has a flat frequency response. Methods and apparatus are presented herein to enhance the acoustic quality of a wideband-to-narrowband converted signal. A bandwidth switching filter is used to emphasize a mid-range frequency portion of the wideband signal so that the resulting narrowband signal has a non-flat frequency spectrum.

Abstract: A sound reproduction system has been developed, for converting signals on two input channels into surround signals on five or seven output channels and vice-versa. A decoder is included in the sound reproduction system which enhances the correlated component of the input signals in the desired direction and reduces the strength of such signals in channels not associated with the encoded direction, while preserving the apparent loudness of all output channels, the separation between the respective left and right output channels and the total energy of the uncorrelated component of the input channels in each output channel. The decoder may include a uniquely defined matrix that helps to ensure that the surface of the output signals is smooth and continuous.

Abstract: The invention relates to a method for suppressing noise interference with the following steps: Gaining of an analytical signal from an input signal (Sin); Calculation of an instant amplitude signal (IA) from the analytical signal; Calculation of an instant phase signal (IFI) from the analytical signal; Non-linear modification of the instant amplitude signal (IA) into a modified instant amplitude signal (IAmod); Linkage of the modified instant amplitude signal (IAmod) with the instant phase signal (IFI) into an output signal (Sout).

Abstract: Extracting features from signals for use in classification, retrieval, or identification of data represented by those signals uses a “Distortion Discriminant Analysis” (DDA) of a set of training signals to define parameters of a signal feature extractor. The signal feature extractor takes signals having one or more dimensions with a temporal or spatial structure, applies an oriented principal component analysis (OPCA) to limited regions of the signal, aggregates the output of multiple OPCAs that are spatially or temporally adjacent, and applies OPCA to the aggregate. The steps of aggregating adjacent OPCA outputs and applying OPCA to the aggregated values are performed one or more times for extracting low-dimensional noise-robust features from signals, including audio signals, images, video data, or any other time or frequency domain signal. Such extracted features are useful for many tasks, including automatic authentication or identification of particular signals, or particular elements within such signals.

Abstract: A portion of an audio signal is separated into multiple frames from which one or more different features are extracted. These different features are used, in combination with a set of rules, to classify the portion of the audio signal into one of multiple different classifications (for example, speech, non-speech, music, environment sound, silence, etc.). In one embodiment, these different features include one or more of line spectrum pairs (LSPs), a noise frame ratio, periodicity of particular bands, spectrum flux features, and energy distribution in one or more of the bands. The line spectrum pairs are also optionally used to segment the audio signal, identifying audio classification changes as well as speaker changes when the audio signal is speech.

Abstract: A system and method for the concealment of errors resulting from missing or corrupted data in the transmission of audio signals in compressed digital packet formats is disclosed. The system utilizes a circular FIFO buffer to store audio frames from the transmitted audio signal, and a beat detector, to identify the presence of beats in the audio signal. The error concealment method replaces erroneous audio frames with error-free audio frames by a process which takes into account the presence and location of the detected beats.

Abstract: An audio decoding apparatus decodes high frequency component signals using a band expander that generates multiple high frequency subband signals from low frequency subband signals divided into multiple subbands and transmitted high frequency encoded information. The apparatus is provided with an aliasing detector and an aliasing remover. The aliasing detector detects the degree of occurrence of aliasing components in the multiple high frequency subband signals generated by the band expander. The aliasing remover suppresses aliasing components in the high frequency subband signals by adjusting the gain used to generate the high frequency subband signals. Thus occurrence of aliasing can be suppressed and the resulting degradation in sound quality can be reduced, even when real-valued subband signals are used in order to reduce the number of operations.

Abstract: A method of speech recognition with compensation is provided by modifying HMM models trained on clean speech with cepstral mean normalization. For all speech utterances the MFCC vector is calculated for the clean database. This mean MFCC vector is added to the original models. An estimate of the background noise is determined for a given speech utterance. The model mean vectors adapted to the noise are determined. The mean vector of the noisy data over the noisy speech space is determined and this is removed from model mean vectors adapted to noise to get the target model.

Abstract: A wideband, high quality audio signal is decoded with few calculations at a low bitrate. Unwanted spectrum components accompanying sinusoidal signal injection by a synthesis subband filter built with real-value operations are suppressed by inserting a suppression signal to subbands adjacent to the subband to which the sine wave is injected. This makes it possible to inject a desired sinusoid with few calculations.

Abstract: In a speech signal decoding method, information containing at least a sound source signal, gain, and filter coefficients is decoded from a received bit stream. Voiced speech and unvoiced speech of a speech signal are identified using the decoded information. Smoothing processing based on the decoded information is performed for at least either one of the decoded gain and decoded filter coefficients in the unvoiced speech. The speech signal is decoded by driving a filter having the decoded filter coefficients by an excitation signal obtained by multiplying the decoded sound source signal by the decoded gain using the result of the smoothing processing. A speech signal decoding apparatus is also disclosed.