Abstract: Speaker module housing and speaker module, the speaker module housing includes front shell with accommodating cavity and back cover fixed to front shell and covering accommodating cavity. Front shell has gate region for injection molding and fixed region opposite to gate region and away from gate region. Thickness of part of front shell between gate region and fixed region is smaller than thickness of fixed region. Speaker module housing further includes plastic structure fixed in accommodating cavity. Plastic structure and front shell are integrally injection-molded, and plastic structure is located in gate region and/or fixed region. Plastic structure is arranged in gate region and/or fixed region of front shell, so that part of front shell located in gate region and fixed region has sufficient plastic thickness, which extends effective holding time of shrinkage position of front shell and reduce risk of shrinkage in fixed region of front shell.

Abstract: Aspects of the subject technology relate to providing device-independent audio for electronic devices. In one or more implementations, microphone data captured by multiple microphones at an electronic device may be provided to a device-specific audio generalizer at the electronic device. The device-specific audio generalizer may utilize device specific information to generalize the microphone data to form device-independent audio data. The device-independent audio data may then be provided to a device-independent machine learning model at the electronic device or another electronic device for further processing.

Abstract: A backpropagation enabled model is trained for estimating time-lapse property changes of a subsurface volume. Synthetic models of the subsurface volume are generated, with pre-determined property changes before and after a time lapse. These models are used to compute baseline-monitor pairs of synthetic seismic traces before and after the time lapse, wherein the baseline synthetic traces are computed from the synthetic model before the time lapse and the monitor synthetic traces are computed from the synthetic model after the time lapse. A ground truth 4D attribute characterizing the time-lapse property changes in the synthetic models is defined, and a backpropagation enabled model is trained by feeding the baseline-monitor pairs of synthetic seismic traces and the corresponding ground truth 4D attribute. The thus obtained trained backpropagation enabled model can be used to estimate time-lapse property changes of the actual subsurface Earth volume from actual baseline-monitor pairs of seismic traces.