Microsoft Technology Licensing, LLC

Ziyad Ibrahim

Robert Norman Heitkamp

Philip Andrew Edry

Steven Wilssens

Brian C McDowell

Paul J Radek

Yongjun Wu

Alexander Robert Paul Grenier

Glenn Frederick Evans

Shyam Sadhwani

Sudhakar Visweswara Prabhu

Steven Marcel Elza Wilssens

Laxmi Narsimha Rao Kakulamarri

Chad R Heinemann

James David McCaffrey

Dharmanshu Kamra

Daniel James Carpenter

Nicholas Fredrick Ray

Andrew Yi Jiang

Technologies are disclosed for improving the efficiency of real-time audio processing, and specifically for improving the efficiency of continuously modifying a real-time audio signal. Efficiency is improved by reducing memory bandwidth requirements and by reducing the amount of processing used to modify the real-time audio signal. In some configurations, memory bandwidth requirements are reduced by selectively transferring active samples in the frequency domain&#x2014;e.g. avoiding the transfer samples with amplitudes of zero or near-zero. This has particular importance when the specialized hardware retrieves samples from main memory in real-time. In some configurations, the amount of processing needed to modify the audio signal is reduced by omitting operations that do not meaningfully affect the output audio signal. For example, a multiplication of samples may be avoided when at least one of the samples has an amplitude of zero or near-zero.

Adaptive coefficients and samples elimination for circular convolution

Examples described herein generally relate to processing event logs where, for each of multiple events in an event log of the one or more event logs, a table of logged event instances can be generated for the event. For each of the multiple events, the table can be processed using an autoencoder to identify one or more of the logged event instances as anomalies, and an indication of at least a portion of the anomalies can be output. In addition, the event logs and/or corresponding tables of events can be used to train models for the autoencoders.

Techniques for detecting atypical events in event logs

Examples are disclosed that relate to managing workloads provided by a plurality of clients to a shared resource. One example provides a computing device comprising a processor and a storage device storing instructions executable by the processor. The instructions are executable to provide a first workload from a first client and a second workload from a second client to a shared memory accessible by the first client, the second client, and a resource configured to process the first workload and the second workload. The computing device is configured to determine that an exception has occurred while processing the first workload, and to take an action to prevent execution of at least some additional work from the first client. The instructions are further executable to receive a result of processing the second workload after taking the action to prevent the execution of the additional work.

Managing exceptions on a shared resource

An electronic display system comprises first and second display surfaces and a computer. Each of the first and second display surfaces is configured to receive and transmit display light from an emissive element. Each of the first and second display surfaces includes both a flat portion and an edge portion non-coplanar to the flat portion. The computer is configured to control the emissive elements of the first and second display surfaces so as to present a first section of a display image on the first display surface and a second section of the display image on the second display surface. In this example, one or more rows of the display image rendered on the flat portion of a display surface are rendered duplicatively on an edge portion.

Display presentation across plural display surfaces

The techniques disclosed herein provide apparatuses and related methods for the communication of spatial audio and related metadata. In some implementations, a source provides prerecorded spatial audio that has embedded metadata. A computing device processes the prerecorded spatial audio to generate an audio codec that is segmented to include a first section of audio data and a second section that includes metadata extracted from the prerecorded spatial audio. The generated audio codec may be received by a device that includes an encoder. The encoder may process the generated audio codec to generate audio data that includes the metadata.

Frame coding for spatial audio data

A system for enabling a shared three-dimensional ('3D') audio bed available to multiple software applications is provided. The system manages bed metadata defining a number of speaker objects of a 3D audio bed. The bed metadata also associates each speaker object with a location, which in some configurations, is defined by a three-dimensional coordinate system. The bed metadata is communicated to a plurality of applications. The applications can then generate custom 3D audio data that associates individual audio streams with individual speaker objects of the 3D audio bed. The applications can then communicate the custom 3D audio data to a 3D audio bed engine, which causes the processing and rendering of the custom 3D audio data to an output device utilizing a selected spatialization technology. Aspects of the 3D bed can be altered when the spatialization technology or the output device changes.

Shared three-dimensional audio bed

The techniques disclosed herein can enable a system to coordinate the processing of object-based audio and channel-based audio generated by multiple applications. The system determines a spatialization technology to utilize based on contextual data. In some configurations, the contextual data can indicate the capabilities of one or more computing resources. In some configurations, the contextual data can also indicate preferences. The preferences, for example, can indicate user preferences for a type of spatialization technology, e.g., Dolby Atmos, over another type of spatialization technology, e.g., DTSX. Based on the contextual data, the system can select a spatialization technology and a corresponding encoder to process the input signals to generate a spatially encoded stream that appropriately renders the audio of multiple applications to an available output device. The techniques disclosed herein also allow a system to dynamically change the spatialization technologies during use.

Adaptive audio rendering

Innovations in the area of sample metadata processing can help a media playback tool avoid loss of synchronization between sample metadata and media samples. For example, a media playback tool identifies encoded data and sample metadata for a current media sample, then couples the sample metadata with the current media sample. The media playback tool provides the sample metadata and encoded data for the current media sample to a media decoder, which maintains the coupling between at least one element of the sample metadata and the current media sample during at least one stage of decoding, even when the current media sample is dropped, delayed, split, or repeated. For example, the media playback tool can determine whether to drop the current media sample and, if the current media sample is dropped, also drop the sample metadata that is coupled with the current media sample.

Coupling sample metadata with media samples

The present disclosure enables applications of a computing system to coordinate object-based audio resources by the use of a minimum resource working set. The minimum resource working set encourages an application to be fair in its requirements since specifying a large number will most likely result in the application receiving zero resources, or losing all of its resources to another application. A working set, which can include a minimum and a maximum working set, also provides a useful metric for the spatial audio resource manager to use when balancing demand. In addition, a minimum working set provides a performance metric for resource balancing since it exposes what the minimum functional requirement is from the maxim requested resource claim.

Spatial audio resource management utilizing minimum resource working sets

Technologies for a single-pass/single copy network abstraction layer unit ('NALU') parser. Such a NALU parser typically reuses source and/or destination buffers, optionally changes endianess of NALU data, optionally processes emulation prevention codes, and optionally processes parameters in slice NALUs, all as part of a single pass/single copy process. The disclosed NALU parser technologies are further suitable for hardware implementation, software implementation, or any combination of the two.

Growing community of inventors

Redmond, WA, United States of America

Ziyad Ibrahim