Abstract: A device includes a memory configured to store one or more segments of time-series data. The device also includes one or more processors configured to generate, using a feature extractor, a latent-space representation of a segment of the time-series data. The one or more processors are also configured to provide one or more inputs to a classifier, the one or more inputs including at least one input based on the latent-space representation. The one or more processors are also configured to generate, based on output of the classifier, a processing control signal for the segment.

Abstract: A device includes one or more processors configured to input one or more segments of an input media stream into a feature extractor. The one or more processors are further configured to pass an output of the feature extractor into an utterance classifier to produce at least one representation of at least one utterance class of a plurality of utterance classes. The one or more processors are further configured to pass the output of the feature extractor and the at least one representation into a segment matcher to produce a media output segment identifier.

Abstract: Systems and techniques are described herein for encoding and/or decoding audio information. For example, a process can process an input audio segment to generate a representation of the input audio segment, and can compare the representation of the input audio segment to representations stored in a memory. The representations represent a plurality of audio segments. The process can determine, based on the comparison, target representation(s) of target audio segment(s) from the representations stored in the memory. The process can determine one or more indices associated with the target audio segment(s). The process can then packetize the one or more indices and transmit the one or more packetized indices (e.g., to a decoder configured to decode the packetized indices).

Abstract: A device includes one or more processors configured to input one or more segments of an input media stream into a feature extractor. The one or more processors are further configured to pass an output of the feature extractor into an utterance classifier to produce at least one representation of at least one utterance class of a plurality of utterance classes. The one or more processors are further configured to pass the output of the feature extractor and the at least one representation into a segment matcher to produce a media output segment identifier.

Abstract: A method includes extracting a voicing classification parameter of an audio signal and determining a filter coefficient of a low pass filter based on the voicing classification parameter. The method also includes filtering a low-band portion of the audio signal to generate a low-band audio signal and controlling an amplitude of a temporal envelope of the low-band audio signal based on the filter coefficient. The method also includes modulating a white noise signal based on the amplitude of the temporal envelope to generate a modulated white noise signal and scaling the modulated white noise signal based on a noise gain to generate a scaled modulated white noise signal. The method also includes mixing a scaled version of the low-band audio signal with the scaled modulated white noise signal to generate a high-band excitation signal that is used to generate a decoded version of the audio signal.

Abstract: A neural network processing system be configured to: (a) execute a first neural network and a second neural network; (b) run a first data segment through the first neural network to return a first score and run a second data segment through the second neural network to return a second score; (c) compare the first score with the second score; and (d) retrain the first neural network based on the comparison

Abstract: A device includes a decoder configured to receive an encoded audio signal at a decoder and to generate a synthesized signal based on the encoded audio signal. The device further includes a classifier configured to classify the synthesized signal based on at least one parameter determined from the encoded audio signal.

Abstract: A method includes extracting a voicing classification parameter of an audio signal and determining a filter coefficient of a low pass filter based on the voicing classification parameter. The method also includes filtering a low-band portion of the audio signal to generate a low-band audio signal and controlling an amplitude of a temporal envelope of the low-band audio signal based on the filter coefficient. The method also includes modulating a white noise signal based on the amplitude of the temporal envelope to generate a modulated white noise signal and scaling the modulated white noise signal based on a noise gain to generate a scaled modulated white noise signal. The method also includes mixing a scaled version of the low-band audio signal with the scaled modulated white noise signal to generate a high-band excitation signal that is used to generate a decoded version of the audio signal.

Abstract: A particular method includes determining, at a device, a voicing classification of an input signal. The input signal corresponds to an audio signal. The method also includes controlling an amount of an envelope of a representation of the input signal based on the voicing classification. The method further includes modulating a white noise signal based on the controlled amount of the envelope. The method also includes generating a high band excitation signal based on the modulated white noise signal.

Abstract: A device includes a decoder configured to receive an encoded audio signal at a decoder and to generate a synthesized signal based on the encoded audio signal. The device further includes a classifier configured to classify the synthesized signal based on at least one parameter determined from the encoded audio signal.

Abstract: Systems and methods of performing blind bandwidth extension are disclosed. In an embodiment, a method includes determining, based on a set of low-band parameters of an audio signal, a first set of high-band parameters and a second set of high-band parameters. The method further includes generating a predicted set of high-band parameters based on a weighted combination of the first set of high-band parameters and the second set of high-band parameters.

Abstract: A particular method includes determining, at a device, a voicing classification of an input signal. The input signal corresponds to an audio signal. The method also includes controlling an amount of an envelope of a representation of the input signal based on the voicing classification. The method further includes modulating a white noise signal based on the controlled amount of the envelope. The method also includes generating a high band excitation signal based on the modulated white noise signal.

Abstract: Systems and methods of performing blind bandwidth extension are disclosed. In an embodiment, a method includes determining, based on a set of low-band parameters of an audio signal, a first set of high-band parameters and a second set of high-band parameters. The method further includes generating a predicted set of high-band parameters based on a weighted combination of the first set of high-band parameters and the second set of high-band parameters.

Abstract: Systems and methods of performing blind bandwidth extension are disclosed. In an embodiment, a method includes receiving, at a decoder of a speech vocoder, a set of low-band parameters as part of a narrowband bitstream. The set of low-band parameters are received from an encoder of the speech vocoder. The method also includes predicting a set of high-band parameters based on the set of low-band parameters.