Abstract: Described herein is a method for Convolutional Neural Network (CNN) based speech source separation, wherein the method includes the steps of: (a) providing multiple frames of a time-frequency transform of an original noisy speech signal; (b) inputting the time-frequency transform of said multiple frames into an aggregated multi-scale CNN having a plurality of parallel convolution paths; (c) extracting and outputting, by each parallel convolution path, features from the input time-frequency transform of said multiple frames; (d) obtaining an aggregated output of the outputs of the parallel convolution paths; and (e) generating an output mask for extracting speech from the original noisy speech signal based on the aggregated output. Described herein are further an apparatus for CNN based speech source separation 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: The present application relates to a method of extracting audio features in a dialog detector in response to an input audio signal, the method comprising dividing the input audio signal into a plurality of frames, extracting frame audio features from each frame, determining a set of context windows, each context window including a number of frames surrounding a current frame, deriving, for each context window, a relevant context audio feature for the current frame based on the frame audio features of the frames in each respective context, and concatenating each context audio feature to form a combined feature vector to represent the current frame. The context windows with the different length can improve the response speed and improve robustness.

Abstract: Example embodiments disclosed herein relate to audio object clustering. A method for metadata-preserved audio object clustering is disclosed. The method comprises classifying a plurality of audio objects into a number of categories based on information to be preserved in metadata associated with the plurality of audio objects. The method further comprises assigning a predetermined number of clusters to the categories and allocating an audio object in each of the categories to at least one of the clusters according to the assigning. Corresponding system and computer program product are also disclosed.

Abstract: A method of processing audio content including a plurality of audio elements comprises: clustering the plurality of audio elements into a plurality of clusters of audio elements; and for a cluster among the plurality of clusters: for each audio element in the cluster, determining a measure of energy that the audio element contributes to the cluster; for at least one audio element in the cluster, determining a compensation gain based at least in part on the measures of energy for the audio elements in the cluster; and applying the compensation gain to the at least one audio element in the cluster.

Abstract: Volume leveler controller and controlling method are disclosed. In one embodiment, A volume leveler controller includes an audio content classifier for identifying the content type of an audio signal in real time; and an adjusting unit for adjusting a volume leveler in a continuous manner based on the content type as identified. The adjusting unit may configured to positively correlate the dynamic gain of the volume leveler with informative content types of the audio signal, and negatively correlate the dynamic gain of the volume leveler with interfering content types of the audio signal.

Abstract: An apparatus and method of blind detection of binauralized audio. If the input content is detected as binaural, a second binauralization may be avoided. In this manner, the user experience avoids audio artifacts introduced by multiple binauralizations.

Abstract: Diffuse or spatially large audio objects may be identified for special processing. A decorrelation process may be performed on audio signals corresponding to the large audio objects to produce decorrelated large audio object audio signals. These decorrelated large audio object audio signals may be associated with object locations, which may be stationary or time-varying locations. For example, the decorrelated large audio object audio signals may be rendered to virtual or actual speaker locations. The output of such a rendering process may be input to a scene simplification process. The decorrelation, associating and/or scene simplification processes may be performed prior to a process of encoding the audio data.

Abstract: Volume leveler controller and controlling method are disclosed. In one embodiment, A volume leveler controller includes an audio content classifier for identifying the content type of an audio signal in real time; and an adjusting unit for adjusting a volume leveler in a continuous manner based on the content type as identified. The adjusting unit may configured to positively correlate the dynamic gain of the volume leveler with informative content types of the audio signal, and negatively correlate the dynamic gain of the volume leveler with interfering content types of the audio signal.

Abstract: Described herein is a method for creating object-based audio content from a text input for use in audio books and/or audio play, the method including the steps of: a) receiving the text input; b) performing a semantic analysis of the received text input; c) synthesizing speech and effects based on one or more results of the semantic analysis to generate one or more audio objects; d) generating metadata for the one or more audio objects; and e) creating the object-based audio content including the one or more audio objects and the metadata. Described herein are further a computer-based system including one or more processors configured to perform said method and a 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: Embodiments of training signal processing models for component replacement in signal processing systems are disclosed. A device for training a third signal processing model via a machine learning method include a first calculating unit, a second calculating unit and a training unit. The first calculating unit calculates a first output from each first sample in a first sample set based on a first signal processing model. The second calculating unit calculates a second output from the first sample based on a second signal processing model. The training unit trains a third signal processing model by minimizing a first cost including a first error between the first output and a third output, so that the combination of the second signal processing model and the third signal processing model can simulate the behaviors of the first signal processing model on the first sample set. The third output is an output of the third signal processing model in response to an input including the second output.

Abstract: Described herein is a method for creating object-based audio content from a text input for use in audio books and/or audio play, the method including the steps of: a) receiving the text input; b) performing a semantic analysis of the received text input; c) synthesizing speech and effects based on one or more results of the semantic analysis to generate one or more audio objects; d) generating metadata for the one or more audio objects; and e) creating the object-based audio content including the one or more audio objects and the metadata. Described herein are further a computer-based system including one or more processors configured to perform said method and a 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: Diffuse or spatially large audio objects may be identified for special processing. A decorrelation process may be performed on audio signals corresponding to the large audio objects to produce decorrelated large audio object audio signals. These decorrelated large audio object audio signals may be associated with object locations, which may be stationary or time-varying locations. For example, the decorrelated large audio object audio signals may be rendered to virtual or actual speaker locations. The output of such a rendering process may be input to a scene simplification process. The decorrelation, associating and/or scene simplification processes may be performed prior to a process of encoding the audio data.

Abstract: An apparatus and method of blind detection of binauralized audio. If the input content is detected as binaural, a second binauralization may be avoided. In this manner, the user experience avoids audio artifacts introduced by multiple binauralizations.

Abstract: Example embodiments disclosed herein relate to audio source separation with source direction determined based on iterative weighted component analysis. A method of separating audio sources in audio content is disclosed. The audio content includes a plurality of channels. The method includes obtaining multiple data samples from multiple time-frequency tiles of the audio content. The method also includes analyzing the data samples to generate multiple components in a plurality of iterations, wherein each of the components indicates a direction with a variance of the data samples, and wherein in each of the plurality of iterations, each of the data samples is weighted with a weight that is determined based on a selected component from the multiple components. The method further includes determining a source direction of the audio content based on the selected component for separating an audio source from the audio content.

Abstract: Example embodiments disclosed herein relate to signal processing. A method for decomposing a plurality of audio signals from at least two different channels is disclosed. The method comprises obtaining a set of components that are weakly correlated, the set of components generated based on the plurality of audio signals. The method comprises extracting a feature from the set of components, and determining a set of gains associated with the set of components at least in part based on the extracted feature, each of the gains indicating a proportion of a diffuse part in the associated component. The method further comprises decomposing the plurality of audio signals by applying the set of gains to the set of components. Corresponding system and computer program product are also disclosed.

Abstract: Volume leveler controller and controlling method are disclosed. In one embodiment, A volume leveler controller includes an audio content classifier for identifying the content type of an audio signal in real time; and an adjusting unit for adjusting a volume leveler in a continuous manner based on the content type as identified. The adjusting unit may configured to positively correlate the dynamic gain of the volume leveler with informative content types of the audio signal, and negatively correlate the dynamic gain of the volume leveler with interfering content types of the audio signal.

Abstract: The present document describes a method for extracting J audio sources from I audio channels. The method includes updating a Wiener filter matrix based on a mixing matrix from a source matrix and based on a power matrix of the J audio sources. Furthermore, the method includes updating a cross-covariance matrix of the I audio channels and of the J audio sources and an auto-covariance matrix of the J audio sources, based on the updated Wiener filter matrix and based on an auto-covariance matrix of the I audio channels. In addition, the method includes updating the mixing matrix and the power matrix based on the updated cross-covariance matrix of the I audio channels and of the J audio sources, and/or based on the updated auto-covariance matrix of the J audio sources.

Abstract: Apparatus and methods for audio classifying and processing are disclosed. In one embodiment, an audio processing apparatus includes an audio classifier for classifying an audio signal into at least one audio type in real time; an audio improving device for improving experience of audience; and an adjusting unit for adjusting at least one parameter of the audio improving device in a continuous manner based on the confidence value of the at least one audio type.

Abstract: Example embodiments disclosed herein relate to audio object clustering based on renderer-aware perceptual difference. A method of processing audio objects is provided. The method includes obtaining renderer-related information indicating a configuration of a renderer. The method also includes determining, based on the obtained renderer-related information, a rendering difference between a first audio object and a second audio object among the audio objects with respect to the renderer. The method further includes clustering the audio objects at least in part based on the rendering difference. Corresponding system, device, and computer program product are also disclosed.

Abstract: An audio processing system and method which calculates, based on spatial metadata of the audio object, a panning coefficient for each of the audio objects in relation to each of a plurality of predefined channel coverage zones. Converts the audio signal into submixes in relation to the predefined channel coverage zones based on the calculated panning coefficients and the audio objects. Each of the submixes indicating a sum of components of the plurality of the audio objects in relation to one of the predefined channel coverage zones. Generating a submix gain by applying an audio processing to each of the submix and controls an object gain applied to each of the audio objects. The object gain being as a function of the panning coefficients for each of the audio objects and the submix gains in relation to each of the predefined channel coverage zones.