Abstract: Implementations described herein relate to providing noise suppression for speech data with reduced power consumption. In some implementations, a computer-implemented method includes receiving a current time frame of speech data, e.g., after receiving a previous time frame associated with a previous noise suppression mask. The current time frame is transformed to a current frequency frame in the frequency domain. A noise classifier is used to determine whether to create a current noise suppression mask for the current frame. If it is determined to create the mask, the mask is created and multiplied by the current frequency frame to obtain a noise-suppressed frequency frame. If it is determined to not create the current mask, the previous noise suppression mask is multiplied with the current frequency frame to obtain the noise-suppressed frequency frame, without creating a mask. The noise-suppressed frequency frame is transformed to a time frame and output.

Abstract: This document generally relates to techniques for testing or training data augmentation. One example includes a method or technique that can include accessing a repository of private data items. The repository can provide a distribution of the private data items that is representative of a designated real-world scenario for a machine learning model. The method or technique can also include assigning classifications to the private data items in the repository. The method or technique can also include augmenting a testing or training set for the machine learning model based at least on the classifications of the private data items to obtain an augmented testing or training set that is relatively more representative of the distribution of classifications in the repository.

Abstract: This document relates to training and employing of quality estimation models to estimate the quality of different signal characteristics. One example includes a method or technique that can be performed on a computing device. The method or technique can include obtaining training signals exhibiting diverse impairments introduced when the training signals are captured or diverse artifacts introduced by different processing characteristics of a plurality of data enhancement models. The method or technique can also include obtaining quality labels for different signal characteristics of the training signals. The method or technique can also include training at least two different quality estimation models to estimate quality of at least two different signal characteristics based at least on the training signals and the quality labels.

Abstract: This document relates to training and employing a quality estimation model. One example includes a method or technique that can be performed on a computing device. The method or technique can include obtaining training signals exhibiting diverse impairments introduced when the training signals are captured or diverse artifacts introduced by different processing characteristics of a plurality of data enhancement models. The method or technique can also include obtaining quality labels for the training signals, and training a quality estimation model to estimate signal quality based at least on the training signals and the quality labels.