Abstract: A method for constructing a decoding network, a speech recognition method, a device, an apparatus, and a storage medium are provided. The method for constructing a decoding network includes: acquiring a general language model, a domain language model, and a general decoding network generated based on the general language model; generating a domain decoding network based on the domain language model and the general language model; and integrating the domain decoding network with the general decoding network to obtain a target decoding network. The speech recognition method includes: decoding to-be-recognized speech data by using a target decoding network to obtain a decoding path for the to-be-recognized speech data; and determining a speech recognition result for the to-be-recognized speech data based on the decoding path for the to-be-recognized speech data.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for utilizing multiple modalities to generate accurate two-dimensional floorplans based on sparse digital videos depicting three-dimensional space. In particular, in one or more embodiments, the disclosed systems extract both visual and audio information from sparse digital video coverage of portions of a three-dimensional space and utilize the extracted visual and audio information to generate a two-dimensional floorplan representing both viewed and unviewed portions of the three-dimensional space. For example, the disclosed systems utilize self-attention layers of a specialized machine learning model to maintain and leverage bi-directional relationships among sequences of visual and audio features to generate floorplan predictions associated with the three-dimensional space.

Abstract: An image compression method includes segmenting, by a processor, an image file to be compressed into a plurality of blocks; performing, by the processor, a first pixel conversion based on pixel difference values between blocks from among the plurality of blocks; and storing the image file of which the first pixel conversion is completed in a first format.

Abstract: An apparatus for encoding an audio input signal to obtain an encoded audio signal is provided. The apparatus comprises a transformation module configured to transform the audio input signal from an original domain to a transform domain to obtain a transformed audio signal. Moreover, the apparatus comprises an encoding module, configured to quantize the transformed audio signal to obtain a quantized signal, and configured to encode the quantized signal to obtain the encoded audio signal. The transformation module is configured to transform the audio input signal depending on a plurality of predefined power values of quantization noise in the original domain.

Abstract: Encoding a sequence of digital speech samples into a bit stream includes dividing the digital speech samples into frames including N subframes (where N is an integer greater than 1); computing model parameters for the subframes, the model parameters including spectral parameters; and generating a representation of the frame. The representation includes information representing the spectral parameters of P subframes (where P is an integer and P<N) and information identifying the P subframes. The representation excludes information representing the spectral parameters of the N?P subframes not included in the P subframes.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for compressing and decompressing data. In one aspect, a method comprises: processing data using an encoder neural network to generate a latent representation of the data; processing the latent representation of the data using a hyper-encoder neural network to generate a latent representation of an entropy model; generating an entropy encoded representation of the latent representation of the entropy model; generating an entropy encoded representation of the latent representation of the data using the latent representation of the entropy model; and determining a compressed representation of the data from the entropy encoded representations of: (i) the latent representation of the data and (ii) the latent representation of the entropy model used to entropy encode the latent representation of the data.

Abstract: A method for encoding a multi-channel signal and an encoder, where the encoding method includes obtaining a multi-channel signal of a current frame, determining an initial inter-channel time difference (ITD) value of the current frame, controlling, based on characteristic information of the multi-channel signal, a quantity of target frames that are allowed to appear continuously, where the characteristic information includes at least one of a signal-to-noise ratio of the multi-channel signal or a peak feature of cross correlation coefficients of the multi-channel signal, and an ITD value of a previous frame of the target frame is reused as an ITD value of the target frame, determining an ITD value of the current frame based on the initial ITD value and the quantity of target frames allowed to appear continuously, and encoding the multi-channel signal based on the ITD value of the current frame.

Abstract: It is inter alia disclosed an apparatus comprising: a table comprising a plurality of sub vectors, wherein each entry of the table is a subvector and each subvector have vector components which are the same as vector components of one or more basis code vectors; and a further table wherein an entry of the further table comprises a first pointer pointing to a sub vector in the table and a second pointer pointing to a subvector in the table, wherein the first pointer and the second pointer are arranged in the further table such that when vector components of the sub vector pointed to by the first pointer are combined with vector components of the sub vector pointed to by the second pointer a basis code vector is formed.

Abstract: In general, techniques are described for compressing decomposed representations of a sound field. A device comprising one or more processors may be configured to perform the techniques. The one or more processors may be configured to obtain a bitstream comprising a compressed version of a spatial component of a sound field, the spatial component generated by performing a vector based synthesis with respect to a plurality of spherical harmonic coefficients.

Abstract: An encoding apparatus for encoding a vibrotactile signal includes a first transforming unit configured to perform a discrete wavelet transform of the signal, a second transforming unit configured to generate a frequency domain representation of the signal, a psychohaptic model unit configured to generate at least one quantization control signal based on the generated frequency domain representation of the sampled signal and on a predetermined perceptual model based on human haptic perception, a quantization unit configured to quantize wavelet coefficients resulting from the performed discrete wavelet transform and adapted by the quantization control signal, a compression unit configured to compress the quantized wavelet coefficients, and a bitstream generating unit configured to generate a bitstream corresponding to the encoded signal based on the compressed quantized wavelet coefficients.

Abstract: A voice synthesis method designates a target feature of a voice to be synthesized; specifies harmonic frequencies for a plurality of respective harmonic components of the voice and an amplitude spectrum envelope of the voice; specifies a harmonic amplitude distribution of each of the plurality of respective harmonic components based on (i) the target feature, (ii) the amplitude spectrum envelope, and (iii) the harmonic frequency specified for the respective harmonic component, the harmonic amplitude distribution representing a distribution of amplitudes in a unit band with a peak amplitude corresponding to the respective harmonic component; and generates a frequency spectrum of the voice with the target feature based on harmonic amplitude distributions specified for each of the plurality of respective harmonic components and the amplitude spectrum envelope.

Abstract: A voice-activity detector (VAD) system comprises a microphone operable to receive and to process audio inputs from an environment to generate an analog audio input signal. The system further comprises an analog VAD operable to process the analog audio input signal to perform an initial detection of human speech, and operable to send a wake up command to a digital signal processing chain to awaken the digital signal processing chain from a sleep mode, when the analog VAD detects human speech. Further, the system comprises the digital signal processing chain operable to process the analog audio input signal to perform a secondary detection of human speech, and operable to output a signal indicating that human speech is detected, when the digital signal processing chain detects human speech. In one or more embodiments, the secondary detection of human speech is more accurate than the initial detection of human speech.

Abstract: In one embodiment, a computing system may determine, for a pixel of a current subframe of a series of subframes, a compensated target pixel value based on a difference between (1) an aggregated target pixel value of the pixel in the current subframe and previous subframes of the series of subframes and (2) an aggregated actual pixel value of displayed pixel values of the pixel in the previous subframes of the series of subframes. The compensated target pixel value may compensate a quantization error for the pixel displayed in the previously subframes of the series of subframes. The system may access a mask value, corresponding to a pixel position of the pixel, from a dithering mask having a spatial stacking property. The system may determine a quantized pixel value for the pixel of the current subframe by quantizing the compensated target pixel value based on the mask value.

Abstract: A method of parametric coding of a multichannel digital audio signal including coding a signal arising from a channels reduction processing applied to the multichannel signal and coding spatialization information of the multichannel signal. The method includes the following acts: extraction of a plurality of items of spatialization information of the multichannel signal; obtaining at least one representation model of the extracted spatialization information; determination of at least one angle parameter of a model obtained; coding the at least one determined angle parameter so as to code the spatialization information extracted during the coding of spatialization information. Also provided are a method for decoding such a coded signal and corresponding coding and decoding devices.

Abstract: Systems and methods are disclosed for improving data channel design by applying transform domain analytics to more reliably extract user data from a signal. In certain embodiments, an apparatus may comprise a channel circuit configured to receive an analog signal at an input of the channel circuit, and sample the analog signal to obtain a set of signal samples. The channel circuit may further apply a filter configured to perform transform domain analysis to the set of signal samples to generate a first subset of samples, the first subset including fewer transitions and having a higher signal to noise ratio (SNR) than the set of signal samples. The channel circuit may detect first bit transform domain representation values from the first subset, and determine channel bit values encoded in the analog signal based on the set of signal samples and using the first bit transform domain representation values detected from the first subset as side information.

Abstract: Methods and devices are described for lossless bandsplitting and bandjoining of streams of signal samples using allpass filtering. The bandsplitting operation reformats an original stream into two intermediate streams representing even and odd samples of the original stream, and then matrix filters these to provide two output substreams representing higher frequency components and lower frequency components of the original stream. Conversely, the bandjoining operation matrix filters two subband streams to provide two quantised intermediate substreams, and then interleaves the filtered streams to furnish an output stream, such that the intermediate substreams are the even and odd samples of the output stream.

Abstract: An apparatus for decoding an encoded audio signal to obtain a reconstructed audio signal is provided. The apparatus includes a receiving interface, a delay buffer and a sample processor for processing the selected audio signal samples to obtain reconstructed audio signal samples of the reconstructed audio signal. The sample selector is configured to select, if a current frame is received by the receiving interface and if the current frame being received by the receiving interface is not corrupted, the plurality of selected audio signal samples from the audio signal samples being stored in the delay buffer depending on a pitch lag information being included by the current frame.

Abstract: An improved gain-shape vector quantization is achieved by determining a number of bits to be allocated to a gain adjustment- and shape-quantizer for a plurality of combinations of a current bit rate and a first signal property. The bit allocation is derived by using an average of optimal bit allocations for a training data set. A number of bits to the gain adjustment and the shape quantizers for a plurality of combinations of the bit rate and a first signal are pre-calculated, and a table indicating the number of bits to be allocated to the gain adjustment- and the shape-quantizers for a plurality of combinations of the bit rate and a first signal property is created. In this way, the table can be used for achieving an improved bit allocation.

Abstract: A system and method for the creation and management of one or more investment indexes utilizes a computer-based quantitative investment program without the need to examine or publish the actual source or executable code of such program. This may be accomplished by confirmation and certification, on an ongoing basis, that the identical processes and inherent quantitative rules are sustained and unchanged from the initial date the respective Index was first calculated. The system may consist of a remote server and network facility that separately hosts an input file, the proprietary quantitative software, and an output file.

Abstract: A method for encoding a multi-channel signal and an encoder, where the encoding method includes obtaining a multi-channel signal of a current frame, determining an initial inter-channel time difference (ITD) value of the current frame, controlling, based on characteristic information of the multi-channel signal, a quantity of target frames that are allowed to appear continuously, where the characteristic information includes at least one of a signal-to-noise ratio of the multi-channel signal or a peak feature of cross correlation coefficients of the multi-channel signal, and an ITD value of a previous frame of the target frame is reused as an ITD value of the target frame, determining an ITD value of the current frame based on the initial ITD value and the quantity of target frames allowed to appear continuously, and encoding the multi-channel signal based on the ITD value of the current frame.

Abstract: A compander including a module configured to compress a range of amplitudes of a signal in accordance with a companding function derived using a calculus of variations approach and method for deriving the companding function.

Abstract: An audio signal, having first and second regions of frequency spectrum, is coded. Spectral peaks in the first region are encoded by a first coding method. For a segment of the audio signal, a relation between energy of bands in the first and second regions is determined. A relation between the energy of the band in the second region and energy of neighboring bands in the second region is determined. A determination is made whether available bits are sufficient for encoding at least one non-peak segment of the first region and the band in the second region. Responsive to first and second relations fulfilling a respective predetermined criterion and a sufficient number of bits, encoding the band in the second region using a second coding method different from the first coding method, and otherwise, subjecting the band in the second region to BandWidth Extension (BWE) or noise fill.

Abstract: A method for displaying content by an electronic device is provided. The method includes generating converted data by encoding a plurality of screen data according to first clocks, restoring the screen data by decoding the converted data based on second clocks separate from the first clocks, and displaying the restored screen data.

Abstract: Methods and apparatus for performing signal processing. The signal processing comprises demultiplexing input encoded data into data including information for a segment including frames and coefficient information for a coefficient selected in the frames of the segment, and low band encoded data, decoding the low band encoded data to produce a low band signal, selecting a coefficient of a frame to be processed from a plurality of the coefficients based on the data, calculating a high band sub-band power of a high band sub-band signal of each sub-band constituting a high band signal of the frame to be processed based on a low band sub-band signal of each sub-band constituting the low band signal of the frame to be processed and the selected coefficient, and producing the high band signal of the frame to be processed based on the high band sub-band power and the low band sub-band signal.

Abstract: A method for compensating for acoustic influence of a listening room on an acoustic output from an audio system including at least a left and a right loudspeaker, the method comprising determining a left frequency response and a right frequency response, designing left and right compensation filters, and during playback applying the left and right filters to left and right input signals. The method further includes determining mono and side responses and designing mono and side compensation filters, and, during playback, applying the mono compensation filter to a mono signal based on the left and right input signals, and applying the side compensation filter to a side signal based on the left and right input signals. The filters are thus combined to provide left and right output signals which have been left/right filtered and mono/side filtered.

Abstract: A system for processing compressed audio includes a signal enhancer module configured to generate one or more signal treatments. The one or more signal treatments may be generated by the signal enhancer module based on analysis of the incoming audio signal. Alternatively, or in addition, characteristics of the incoming audio signal may be provided to the signal enhancer module for use in generating the one or more signal treatments. The one or more signal treatments may be added to file audio signals.

Abstract: According to an example, a communication device is described comprising a receiver configured to receive a signal, a divider configured to divide the signal into signal components, an estimator configured to estimate, for each signal component, an expected processing error which is made when, instead of a first processing scheme, a second processing scheme is used to process the signal component, wherein the first has a higher processing effort than the second, a determiner configured to determine, for each signal component, whether to process the signal component by the first or by the second processing scheme based on the expected processing errors.

Abstract: A method for compressing IQ data for high speed transport link and an associated device. The method comprises: determining, based on dynamical statistical distribution of the IQ data, one or more parameters of a companding function for a nonlinear companding operation (S310); applying the companding function with the determined one or more parameters on the IQ data (S320); performing uniform quantization on the IQ data to generate compressed IQ data (S330); and transmitting the compressed IQ (S340). And a method for decompressing compressed IQ data for high speed transport link, and an associated device.

Abstract: Methods and apparatus for performing signal processing. The signal processing comprises demultiplexing input encoded data into data including information for a segment including frames and coefficient information for a coefficient selected in the frames of the segment, and low band encoded data, decoding the low band encoded data to produce a low band signal, selecting a coefficient of a frame to be processed from a plurality of the coefficients based on the data, calculating a high band sub-band power of a high band sub-band signal of each sub-band constituting a high band signal of the frame to be processed based on a low band sub-band signal of each sub-band constituting the low band signal of the frame to be processed and the selected coefficient, and producing the high band signal of the frame to be processed based on the high band sub-band power and the low band sub-band signal.

Abstract: Provided is an apparatus and method of encoding and decoding multiple channel signals based upon phase information and one or more residual signals.

Abstract: A method transmits an audio signal from a transmitter to a receiver and a hearing device, particularly a hearing aid, contains a communication facility which is provided and configured for transmitting and/or receiving an audio signal according to the method. A hearing device system has two hearing devices and is provided and configured to transmit audio signals between the two hearing devices by their communication facilities according to the method.

Abstract: It is inter alia disclosed to quantize a vector of a plurality of coefficients using a predictive mode of operation of a vector quantizer, wherein the vector quantizer can operate in either a predictive mode of operation or a non-predictive mode of operation, determine a vector of a plurality of recovered coefficients; compare the vector of the plurality of coefficients to the vector of the plurality of recovered coefficients; and determine the mode of operation of the vector quantizer for a vector of a plurality of coefficients associated with a subsequent frame of audio samples, wherein the mode of operation is dependent on the comparison.

Abstract: The system includes a correlation extraction means for extracting at least one candidate for a correlation from a collected given data set, based on a relationship between units of data in the given data set; a correlation verification means for verifying whether or not the units of data in the given data set satisfy the correlation extracted by the correlation extraction means; and a data compression means for compressing the given data set with use of the correlation, based on the result of verification by the correlation verification means.

Abstract: In an approach for collecting data, a computer identifies a virtual map of an environment. The computer determines a location associated with one or more autonomous devices within the identified virtual map. The computer provides to a first autonomous device of the one or more autonomous devices, navigation instructions from a determined location associated with the first autonomous device to a first training point. The computer collects training data associated with the first training point through the first autonomous device.

Abstract: A method for mitigating potential frame instability by an electronic device is described. The method includes obtaining a frame subsequent in time to an erased frame. The method also includes determining whether the frame is potentially unstable. The method further includes applying a substitute weighting value to generate a stable frame parameter if the frame is potentially unstable.

Abstract: An acoustic signal coding apparatus includes a subband classifier that classifies subbands obtained by dividing a frequency-domain spectrum into a plurality of perceptually important first-category subbands and the other subbands referred to as second-category subbands according to at least one of measures in terms of energy and peak property, a subband peak-algebraic vector quantization (SBP-AVQ) vector generator that generates an SBP-AVQ vector by collecting a maximum peak from each first-category subband, outputs the generated SBP-AVQ vector, and outputs peak position information indicating the positions of the maximum peaks, a bit distributor that distributes bits for AVQ coding to the SBP-AVQ vector and the second-category subband vector, and an AVQ coder that performs AVQ coding on the SBP-AVQ vector and the second-category subband vector.

Abstract: Methods and apparatus relating to electrical margining of multi-parameter high-speed interconnect links with multi-sample probing are described. In one embodiment, logic is provided to generate one or more parameter values, corresponding to an electrical operating margin of an interconnect. The one or more parameter values are generated based on a plurality of eye observation sets to be detected in response to operation of the interconnect in accordance with a plurality of parameter sets (e.g., by using quantitative optimization techniques). In turn, the interconnect is to be operated at the one or more parameter values if it is determined that the one or more parameter values cause the interconnect to operate at an optimum level relative to an operation of the interconnect in accordance with one or more less optimum parameter levels. Other embodiments are also disclosed and claimed.

Abstract: An audio processing system (100) comprises a front-end component (102, 103), which receives quantized spectral components and performs an inverse quantization, yielding a time-domain representation of an intermediate signal. The audio processing system further comprises a frequency-domain processing stage (104, 105, 106, 107, 108), configured to provide a time-domain representation of a processed audio signal, and a sample rate converter (109), providing a reconstructed audio signal sampled at a target sampling frequency. The respective internal sampling rates of the time-domain representation of the intermediate audio signal and of the time-domain representation of the processed audio signal are equal. In particular embodiments, the processing stage comprises a parametric upmix stage which is operable in at least two different modes and is associated with a delay stage that ensures constant total delay.

Abstract: An audio encoding apparatus capable of reducing the bit rate even if a codebook having a larger codebook number is selected in a split multi-rate lattice vector quantization is provided. Sub-vector determining unit (121) determines, in the spectrum of an input signal having been divided into a predetermined number of sub-vectors, a sub-vector using the largest number of bits. Positional information encoding unit (122) encodes the positional information of the determined sub-vector. Codebook indication value estimating unit (124) estimates a number of used bits for a codebook indication value of the largest number of used bits by use of the (N?1) other codebook indication values, and generates a number-of-used-bits estimation value. Difference calculating unit (125) calculates a difference by subtracting the number-of-used-bits estimation value from the actual value of the codebook indication value of the largest number of used bits. Difference encoding unit (126) encodes the difference information.

Abstract: An apparatus for generating a frequency enhancement signal has: a signal generator for generating an enhancement signal from a core signal, the enhancement signal having an enhancement frequency range not included in the core signal, wherein a current time portion of the enhancement signal or the core signal has subband signals for a plurality of subbands; a controller for calculating the same smoothing information for the plurality of subband signals of the enhancement frequency range or the core signal, and wherein the signal generator is configured for smoothing the plurality of subband signals of the enhancement frequency range or the core signal using the same smoothing information.

Abstract: A video coder applies a residual differential pulse code modulation technique to a residual data of a block coded using lossy coding. The block may be coded without application of a transform to the residual data.

Abstract: Method of audio power reduction and thermal mitigation using psychoacoustic techniques starts by receiving a decoded audio signal in a reproduction system. Decoded audio signal is a signal that is decompressed and to be played back by a speaker. A masking curve is generated based on psychoacoustic models and the decoded audio signal. The masking curve is applied to the decoded audio signal to remove unheard frequencies and to generate a power-reduced audio signal. Other embodiments are also described.

Abstract: Methods and apparatus for performing signal processing. The signal processing comprises demultiplexing input encoded data into data including information for a segment including frames and coefficient information for a coefficient selected in the frames of the segment, and low band encoded data, decoding the low band encoded data to produce a low band signal, selecting a coefficient of a frame to be processed from a plurality of the coefficients based on the data, calculating a high band sub-band power of a high band sub-band signal of each sub-band constituting a high band signal of the frame to be processed based on a low band sub-band signal of each sub-band constituting the low band signal of the frame to be processed and the selected coefficient, and producing the high band signal of the frame to be processed based on the high band sub-band power and the low band sub-band signal.

Abstract: An apparatus for coding video information according to certain aspects includes a memory and a processor. The memory unit is configured to store video information associated with a reference layer picture and an enhancement layer picture. The processor is configured to: store video information associated with a reference layer picture and an enhancement layer picture; receive a scale factor that indicates a proportion of scaling between the reference layer picture and the enhancement layer picture in a first direction; determine, without performing a division operation, a rounding offset value using the scale factor; and determine a coordinate in the first direction of a first sample located in the reference layer picture that corresponds to a second sample located in the enhancement layer picture using the scale factor and the rounding offset value.

Abstract: A system for processing compressed audio includes a signal enhancer module configured to generate one or more signal treatments. The one or more signal treatments may be generated by the signal enhancer module based on analysis of the incoming audio signal. Alternatively, or in addition, characteristics of the incoming audio signal may be provided to the signal enhancer module for use in generating the one or more signal treatments. The one or more signal treatments may be added to file audio signals.

Abstract: A method, medium, and apparatus encoding and/or decoding a multichannel audio signal. The method includes detecting the type of spatial extension data included in an encoding result of an audio signal, if the spatial extension data is data indicating a core audio object type related to a technique of encoding core audio data, detecting the core audio object type; decoding core audio data by using a decoding technique according to the detected core audio object type, if the spatial extension data is residual coding data, decoding the residual coding data by using the decoding technique according to the core audio object type, and up-mixing the decoded core audio data by using the decoded residual coding data. According to the method, the core audio data and residual coding data may be decoded by using an identical decoding technique, thereby reducing complexity at the decoding end.

Abstract: An audio signal decoder includes a context-based spectral value decoder configured to decode a codeword describing one or more spectral values or at least a portion of a number representation thereof in dependence on a context state. The audio signal decoder also includes a context state determinator configured to determine a current context state in dependence on one or more previously decoded spectral values and a time warping frequency-domain-to-time-domain converter configured to provide a time-warped time-domain representation of a given audio frame on the basis of a set of decoded spectral values provided by the context-based spectral value decoder and in dependence on the time warp information. The context-state determinator is configured to adapt the determination of the context state to a change of a fundamental frequency between subsequent audio frames. An audio signal encoder applies a comparable concept.

Abstract: A system for processing compressed audio includes a signal enhancer module configured to generate one or more signal treatments. The one or more signal treatments may be generated by the signal enhancer module based on analysis of the incoming audio signal. Alternatively, or in addition, characteristics of the incoming audio signal may be provided to the signal enhancer module for use in generating the one or more signal treatments. The one or more signal treatments may be added to file audio signals.

Abstract: Exemplary embodiments may provide a method of encoding an audio signal. The method includes: segmenting the audio signal into a plurality of frames, wherein each of the frames includes M samples and M is a natural number greater than one; applying a first window, a second window, and at least one third window to the frames, wherein a length of the second window is longer than a length of the first window, and a length of the third window is longer than the length of the first window and shorter than the length of the second window; time-frequency transforming the frames to which the first window, the second window, and the at least one third window have been applied; and generating a bitstream including the time-frequency transformed frames.

Abstract: The document relates to modulated sub-sampled digital filter banks, as well as to methods and systems for the design of such filter banks. In particular, the present document proposes a method and apparatus for the improvement of low delay modulated digital filter banks. The method employs modulation of an asymmetric low-pass prototype filter and a new method for optimizing the coefficients of this filter. Further, a specific design for a 64 channel filter bank using a prototype filter length of 640 coefficients and a system delay of 319 samples is given. The method substantially reduces artifacts due to aliasing emerging from independent modifications of subband signals, for example when using a filter bank as a spectral equalizer. The method is preferably implemented in software, running on a standard PC or a digital signal processor (DSP), but can also be hardcoded on a custom chip.