Starkey Laboratories, Inc.

Amazon Technologies, Inc.

Ritwik Giri

Tao Zhang

Arvindh Krishnaswamy

Mehmet Umut Isik

Karim Helwani

Buye Xu

Jean-Marc Valin

Yan Zhao

Joao Felipe Santos

Fangzhou Cheng

Srikanth Venkata Tenneti

Neerad Dilip Phansalkar

Aparna Pandey

Hassan Sawaf

Andrew Paul Breen

Yaser Al-Onaizan

Frederic Philippe Denis Mustiere

Roberto Barra-Chicote

Marcello Federico

Fred Mustiere

Robert Enyedi

Shrikant Venkataramani

A method, comprising receiving at least one sound at an electronic device. The at least one sound is enhanced for the at least one user based on a compound metric. The compound metric is calculated using at least two sound metrics selected from an engineering metric, a perceptual metric, and a physiological metric. The engineering metric comprises a difference between an output signal and a desired signal. At least one of the perceptual metric and the physiological metric is based at least in part on input sensed from the at least one user in response to the received at least one sound.

Electronic device using a compound metric for sound enhancement

Systems, methods, and apparatuses for selecting a model are described. A method includes receiving a request to perform model selection and evaluating multiple models by generating various metrics for each trained model. These metrics include a forewarning time (how much in advance of a failure an alert can be raised), an event recall metric (how many failure events were alerted to in advance), an event precision metric (ratio of true and false positives, and an area under a receiver operating characteristic (ROC) curve. For each trained model, a weighted harmonic mean is calculated from these metrics. A model is then selected based on the calculated means and a report on the selected model is generated and provided.

Prognostics and health management service

Real-time audio enhancement for a target speaker may be performed. An embedding of a sample of speaker audio is created using a trained neural network that performs voice identification. The embedding is then concatenated with the input features of a trained machine learning model for audio enhancement. The audio enhancement model can recognize and enhance a target speaker's speech in a real-time implementation, as the embedding is in the same feature space of the audio enhancement model.

Real-time target speaker audio enhancement

A system comprises an ear-worn electronic device configured to be worn by a wearer. The ear-worn electronic device comprises a processor and memory coupled to the processor. The memory is configured to store an annoying sound dictionary representative of a plurality of annoying sounds pre-identified by the wearer. A microphone is coupled to the processor and configured to monitor an acoustic environment of the wearer. A speaker or a receiver is coupled to the processor. The processor is configured to identify different background noises present in the acoustic environment, determine which of the background noises correspond to one or more of the plurality of annoying sounds, and attenuate the one or more annoying sounds in an output signal provided to the speaker or receiver.

Ear-worn electronic device incorporating annoyance model driven selective active noise control

A group masked autoencoder may be implemented for anomaly detection. An autoencoder network model may be trained without supervision and applied to output an estimated joint probability distribution of normality for a group of frames of time series data. The estimated joint probability distribution may be used to determine an anomaly score for the time series data. An anomaly may be detected according to the anomaly score and a result that indicates a detected anomaly may be provided.

Group masked autoencoder for anomaly detection

Systems, methods, and apparatuses for detecting anomalies using clusters are described. In some examples, a method includes receiving a request to perform anomaly detection using a plurality of clusters; receiving a data point; determining when the received data point is a part of one of the plurality of clusters utilizing a distance to centers of the one or more clusters, wherein: when the received data point is determined to belong to a normal cluster, assigning the received data point to the determined cluster, updating the cluster, and updating a history for the cluster, when the received data point is determined to belong to an anomalous cluster, raising an anomaly, updating the cluster, and updating a history for the cluster, and when the received data point is determined to not belong to any cluster, raising an anomaly.

Techniques for training and using a machine learning model for estimation of reverberation in a multi-task learning framework are described. According to some embodiments, the multi-task learning framework improves the performance of the machine learning model by estimating the amount of reverberation present in an input audio recording as a secondary task to the primary task of generating a clean speech portion of the input audio recording. In one embodiment, a model architecture is selected that takes a noisy reverberant recording as an input and outputs an estimate of a clean (e.g., de-reverberated) signal, an estimate of noise (e.g., background noise), and an estimate of the reverb only portion, with the secondary task of estimating the reverb only portion acting as a regularizer that improves the machine learning model's performance in enhancing the reverberant (e.g., and noisy) input speech.

Speech enhancement machine learning model for estimation of reverberation in a multi-task learning framework

Audio processing may be performed with a convolutional neural network that includes positional embeddings. Audio data may be received at an audio processing system. A convolutional neural network that concatenates frequency-positional embeddings at an input layer may be used to process the audio data. A result of processing the audio data through the convolutional neural network may be used to perform an audio processing task.

Convolutional neural network with positional embeddings for audio processing

Techniques for the generation of dubbed audio for an audio/video are described. An exemplary approach is to receive a request to generate dubbed speech for an audio/visual file; and in response to the request to: extract speech segments from an audio track of the audio/visual file associated with identified speakers; translate the extracted speech segments into a target language; determine a machine learning model per identified speaker, the trained machine learning models to be used to generate a spoken version of the translated, extracted speech segments based on the identified speaker; generate, per translated, extracted speech segment, a spoken version of the translated, extracted speech segments using a trained machine learning model that corresponds to the identified speaker of the translated, extracted speech segment and prosody information for the extracted speech segments; and replace the extracted speech segments from the audio track of the audio/visual file with the spoken versions spoken version of the translated, extracted speech segments to generate a modified audio track.

Multilingual speech translation with adaptive speech synthesis and adaptive physiognomy

Post-filtering may be performed for ratio masks as part of audio enhancement. Audio data may be received. A machine learning model may be applied to generate gain values for different spectrum bands of the audio data. The gain values may then be modified using an envelope post-filter according to a monotonically increasing function applied to the gain values to produce modified gain values used to generate an enhanced version of the audio data.

Ratio mask post-filtering for audio enhancement

Perceptually guided speech enhancement using deep neural networks

A system and method of determining a filter to cancel feedback signals from input signals in a hearing assistance device includes determining feedback signals for a plurality of feedback paths associated with the device, and determining a model of the plurality of feedback paths, with the model having an invariant portion and a time varying portion. A probable structure of the invariant portion is determined to generate a structural constraint to constrain the plurality of feedback paths, and probability distributions to impose the structural constraint on the invariant portion are determined. During an iterative process, the invariant portion is iteratively determined using the determined probability distributions and the feedback path measurements. A measurement noise variance representative of model mismatch is updated, for each iteration, to reduce a probability of a non-desirable determination of an invariant filter, and the invariant filter is determined in response to a criterion for ending the iterative process being satisfied.

Growing community of inventors

Eden Prairie, MN, United States of America

Ritwik Giri