Abstract: A method for decoding an encoded audio bitstream is disclosed. The method includes receiving the encoded audio bitstream and decoding the audio data to generate a decoded lowband audio signal. The method further includes extracting high frequency reconstruction metadata and filtering the decoded lowband audio signal with an analysis filterbank to generate a filtered lowband audio signal. The method also includes extracting a flag indicating whether either spectral translation or harmonic transposition is to be performed on the audio data and regenerating a highband portion of the audio signal using the filtered lowband audio signal and the high frequency reconstruction metadata in accordance with the flag.

Abstract: A method (600) for decoding an encoded audio signal (102) is described. The encoded audio signal (102) comprises a sequence of frames. Furthermore, the encoded audio signal (102) is indicative of a plurality of different dynamic range control (DRC) profiles for a corresponding plurality of different rendering modes. Different subsets of DRC profiles from the plurality of DRC profiles are comprised within different frames of the sequence of frames, such that two or more frames of the sequence of frames jointly comprise the plurality of DRC profiles.

Abstract: Described is a method of processing position information indicative of an object position of an audio object, wherein the object position is usable for rendering of the audio object, that comprises: obtaining listener orientation information indicative of an orientation of a listener's head; obtaining listener displacement information indicative of a displacement of the listener's head; determining the object position from the position information; modifying the object position based on the listener displacement information by applying a translation to the object position; and further modifying the modified object position based on the listener orientation information. Further described is a corresponding apparatus for processing position information indicative of an object position of an audio object, wherein the object position is usable for rendering of the audio object.

Abstract: A method may involve: receiving direction of arrival (DOA) data corresponding to sound emitted by at least a first smart audio device of the audio environment that includes a first audio transmitter and a first audio receiver, the DOA data corresponding to sound received by at least a second smart audio device of the audio environment that includes a second audio transmitter and a second audio receiver, the DOA data corresponding to sound emitted by at least the second smart audio device and received by at least the first smart audio device; receiving one or more configuration parameters corresponding to the audio environment, to one or more audio devices, or both; and minimizing a cost function based at least in part on the DOA data and the configuration parameter(s), to estimate a position and an orientation of at least the first smart audio device and the second smart audio device.

Abstract: In an embodiment, a method comprises: filtering reference audio content items to separate the reference audio content items into different frequency bands; for each frequency band, extracting a first feature vector from at least a portion of each of the reference audio content items, wherein the first feature vector includes at least one audio characteristic of the reference audio content items; obtaining at least one semantic label from at least a portion of each of the reference audio content items; obtaining a second feature vector consisting of the first feature vectors per frequency band and the at least one semantic label; generating, based on the second feature vector, cluster feature vectors representing centroids of clusters; separating the reference audio content items according to the cluster feature vectors; and computing an average target profile for each cluster based on the reference audio content items in the cluster.

Abstract: Described herein is a method for training a machine learning algorithm. The method may comprise receiving a first input multichannel audio signal. The method may comprise generating, using the machine learning algorithm, an intermediate audio signal based on the first input multichannel audio signal. The method may comprise rendering the intermediate audio signal into a first output multichannel audio signal. Further, the method may comprise improving the machine learning algorithm based on a difference between the first input multichannel audio signal and the first output multichannel audio signal. Described herein are further an apparatus for generating an intermediate audio format from an input multichannel audio signal as well as a respective computer program product comprising a computer-readable storage medium with instructions adapted to carry out said method when executed by a device having processing capability.

Abstract: On the basis of a bitstream (P), an n-channel audio signal (X) is reconstructed by deriving an m-channel core signal (Y) and multichannel coding parameters (a) from the bitstream, where 1?m<n. Also derived from the bitstream are pre-processing dynamic range control, DRC, parameters (DRC2) quantifying an encoder-side dynamic range limiting of the core signal. The n-channel audio signal is obtained by parametric synthesis in accordance with the multichannel coding parameters and while cancelling any encoder-side dynamic range limiting based on the pre-processing DRC parameters. In particular embodiments, the reconstruction further includes use of compensated post-processing DRC parameters quantifying a potential decoder-side dynamic range compression. Cancellation of an encoder-side range limitation and range compression are preferably performed by different decoder-side components. Cancellation and compression may be coordinated by a DRC pre-processor.

Abstract: Described herein is a method of processing an audio signal using a deep-learning-based generator, wherein the method includes the steps of: (a) inputting the audio signal into the generator for processing the audio signal; (b) mapping a time segment of the audio signal to a latent feature space representation, using an encoder stage of the generator; (c) upsampling the latent feature space representation using a decoder stage of the generator, wherein at least one layer of the decoder stage applies sinusoidal activation; and (d) obtaining, as an output from the decoder stage of the generator, a processed audio signal. Described are further a method for training said generator and respective apparatus, systems and computer program products.

Abstract: There is provided methods and apparatuses for decoding and encoding of audio signals. In particular, a method for decoding includes receiving a waveform-coded signal having a spectral content corresponding to a subset of the frequency range above a cross-over frequency. The waveform-coded signal is interleaved with a parametric high frequency reconstruction of the audio signal above the cross-over frequency. In this way an improved reconstruction of the high frequency bands of the audio signal is achieved.

Abstract: Described is a method of processing position information indicative of an object position of an audio object, wherein the object position is usable for rendering of the audio object, that comprises: obtaining listener orientation information indicative of an orientation of a listener's head; obtaining listener displacement information indicative of a displacement of the listener's head; determining the object position from the position information; modifying the object position based on the listener displacement information by applying a translation to the object position; and further modifying the modified object position based on the listener orientation information. Further described is a corresponding apparatus for processing position information indicative of an object position of an audio object, wherein the object position is usable for rendering of the audio object.

Abstract: The present disclosure provides a decoder configured to receive a finite bitrate stream that includes a quantized latent frame, where the quantized latent frame includes a quantized representation of a current frame of a signal in a latent domain different from a first domain; to generate a reconstructed latent frame from the quantized latent frame; to use a generative neural network model to perform a task for which the general neural network model has been trained, wherein the task includes to generate parameters for an invertible mapping from the latent domain to the first domain; to reconstruct a current frame of the signal in the first domain, which includes to map the reconstructed latent frame to the first domain by use of the invertible mapping, and to use the reconstructed current frame of the signal in the first domain to update a state of the generative neural network model.

Abstract: Method for encoding scene-based audio is provided. In some implementations, the method involves determining, by an encoder, a spatial direction of a dominant sound component in a frame of an input audio signal. In some implementations, the method involves determining rotation parameters based on the determined spatial direction and a direction preference of a coding scheme to be used to encode the input audio signal. In some implementations, the method involves rotating sound components of the frame based on the rotation parameters such that, after being rotated, the dominant sound component has a spatial direction that aligns with the direction preference of the coding scheme. In some implementations, the method involves encoding the rotated sound components of the frame of the input audio signal using the coding scheme in connection with an indication of the rotation parameters or an indication of the spatial direction of the dominant sound component.

Abstract: In some embodiments, a method, comprises: dividing, using at least one processor, an audio input into speech and non-speech segments; for each frame in each non-speech segment, estimating, using the at least one processor, a time-varying noise spectrum of the non-speech segment; for each frame in each speech segment, estimating, using the at least one processor, speech spectrum of the speech segment; for each frame in each speech segment, identifying one or more non-speech frequency components in the speech spectrum; comparing the one or more non-speech frequency components with one or more corresponding frequency components in a plurality of estimated noise spectra and selecting the estimated noise spectrum from the plurality of estimated noise spectra based on a result of the comparing.

Abstract: An electronic device for encoding a picture is described. The electronic device includes a processor and instructions stored in memory that are in electronic communication with the processor. The instructions are executable to encode a step-wise temporal sub-layer access (STSA) sample grouping. The instructions are further executable to send and/or store the STSA sample grouping.

Abstract: The present document describes a method (700) for encoding a multi-channel input signal (201). The method (700) comprises determining (701) a plurality of downmix channel signals (203) from the multi-channel input signal (201) and performing (702) energy compaction of the plurality of downmix channel signals (203) to provide a plurality of compacted channel signals (404). Furthermore, the method (700) comprises determining (703) joint coding metadata (205) based on the plurality of compacted channel signals (404) and based on the multi-channel input signal (201), wherein the joint coding metadata (205) is such that it allows upmixing of the plurality of compacted channel signals (404) to an approximation of the multi-channel input signal (201). In addition, the method (700) comprises encoding (704) the plurality of compacted channel signals (404) and the joint coding metadata (205).

Abstract: Described is a method of training a deep-learning-based system for sound source separation. The system comprises a separation stage for frame-wise extraction of representations of sound sources from a representation of an audio signal, and a clustering stage for generating, for each frame, a vector indicative of an assignment permutation of extracted frames of representations of sound sources to respective sound sources. The representation of the audio signal is a waveform-based representation. The separation stage is trained using frame-level permutation invariant training. Further, the clustering stage is trained to generate embedding vectors for the frames of the audio signal that allow to determine estimates of respective assignment permutations between extracted sound signals and labels of sound sources that had been used for the frames. Also described is a method of using the deep-learning-based system for sound source separation.

Abstract: A method for decoding an encoded audio bitstream is disclosed. The method includes receiving the encoded audio bitstream and decoding the audio data to generate a decoded lowband audio signal. The method further includes extracting high frequency reconstruction metadata and filtering the decoded lowband audio signal with an analysis filterbank to generate a filtered lowband audio signal. The method also includes extracting a flag indicating whether either spectral translation or harmonic transposition is to be performed on the audio data and regenerating a highband portion of the audio signal using the filtered lowband audio signal and the high frequency reconstruction metadata in accordance with the flag. The high frequency regeneration is performed as a post-processing operation with a delay of 3010 samples per audio channel.

Abstract: Methods for generating an object based audio program which is renderable in a personalizable manner, e.g., to provide an immersive, perception of audio content of the program. Other embodiments include steps of delivering (e.g., broadcasting), decoding, and/or rendering such a program. Rendering of audio objects indicated by the program may provide an immersive experience. The audio content of the program may be indicative of multiple object channels (e.g., object channels indicative of user-selectable and user-configurable objects, and typically also a default set of objects which will be rendered in the absence of a selection by a user) and a bed of speaker channels. Another aspect is an audio processing unit (e.g., encoder or decoder) configured to perform, or which includes a buffer memory which stores at least one frame (or other segment) of an object based audio program (or bitstream thereof) generated in accordance with, any embodiment of the method.

Abstract: In an audio encoder, for audio content received in a source audio format, default gains are generated based on a default dynamic range compression (DRC) curve, and non-default gains are generated for a non-default gain profile. Based on the default gains and non-default gains, differential gains are generated. An audio signal comprising the audio content, the default DRC curve, and differential gains is generated. In an audio decoder, the default DRC curve and the differential gains are identified from the audio signal. Default gains are re-generated based on the default DRC curve. Based on the combination of the re-generated default gains and the differential gains, operations are performed on the audio content extracted from the audio signal.

Abstract: A moving picture coding apparatus 1 includes: a quantization matrix holding unit (112) that holds a quantization matrix (WM) which has already been transmitted in a parameter set and a matrix ID for identifying the quantization matrix (WM), which are associated with each other; and a variable length coding unit (111) that obtains the matrix ID corresponding to the quantization matrix (WM) used for quantization from the quantization matrix holding unit (112) and places the matrix ID in a coded stream Str.