Google Inc.

Rasmus Munk Larsen

Richard Francis Lyon

Benjamin Louis Shaya

Michael Smedegaard

Jeffrey Adgate Dean

Naveen Kumar

Quoc V Le

Samuel Bengio

Azalia Mirhoseini

Mohammad Norouzi

Benoit Steiner

Hieu Hy Pham

Yuefeng Zhou

Jay Yagnik

Justin Richard Wodrich

Ryan Michael Rifkin

Kristen Mangum

Nicholas Matarese

Sarvjeet Singh

Gabriel Fisher Slotnick

Anandsudhakar Kesari

Yi-Fan Chen

Michael B Montvelishsky

Rolando Esparza Palacios

John Roberts Anderson

Che-Yu Kuo

The various implementations described herein include methods, devices, and systems for automatic audio equalization. In one aspect, a method is performed at an electronic device that includes speakers, microphones, processors and memory. The electronic device outputs audio user content from the speakers and automatically equalizes subsequent audio output of the device without user input. The automatic equalization includes: (1) obtaining audio content signals, including receiving outputted audio content at each microphone; (2) determining from the audio content signals phase differences between microphones; (3) obtaining a feature vector based on the phase differences; (4) obtaining a frequency correction from a correction database based on the obtained feature vector; and (5) applying the obtained frequency correction to the subsequent audio output.

Methods and systems for automatically equalizing audio output based on room position

A method for determining a placement for machine learning model operations across multiple hardware devices is described. The method includes receiving data specifying a machine learning model to be placed for distributed processing on multiple hardware devices; generating, from the data, a sequence of operation embeddings, each operation embedding in the sequence characterizing respective operations necessary to perform the processing of the machine learning model; processing the sequence of operation embeddings using a placement recurrent neural network in accordance with first values of a plurality network parameters of the placement recurrent neural network to generate a network output that defines a placement of the operations characterized by the operation embeddings in the sequence across the plurality of devices; and scheduling the machine learning model for processing by the multiple hardware devices by placing the operations on the multiple devices according to the placement defined by the network output.

Device placement optimization with reinforcement learning

The various implementations described herein include methods, devices, and systems for automatic audio equalization. In one aspect, a method is performed at an audio device having one or more processors, memory, and a plurality of device interface elements, including one or more speakers and a plurality of microphones. The method includes: (1) detecting a change in orientation of the audio device from a first orientation to a second orientation; and (2) in response to detecting the change in orientation, configuring operation of two or more of the plurality of device interface elements.

Orientation-based device interface

The various implementations described herein include methods, devices, and systems for automatic audio equalization. In one aspect, a method is performed at a computing system that includes speaker(s), microphones, processors and memory. The computing system outputs audio user content and automatically equalizes the audio output of the computing system. The equalizing includes: (1) receiving the outputted audio content at each microphone of the plurality of microphones; (2) based on the received audio content, determining an acoustic transfer function for the room; (3) based on the determined acoustic transfer function, obtaining a frequency response for the room; and (4) adjusting one or more properties of the speakers based on the determined frequency response.

Methods and systems for automatically equalizing audio output based on room characteristics

Implementations include systems and methods for scoring candidates for set recommendation problems. An example method includes repeating, for each code in code arrays for items in a set of items, determining a most common value for the code. In some implementations, the method includes determining that the most common value occurs with a frequency that meets an occurrence threshold and adding the code and the most common value to set-inclusion criteria. In other implementations, the method includes determining a value for the code from a code array for a seed item and adding the code and the most common value to set-inclusion criteria when the value for the code from the code array for the seed item matches the most common value. The method may also include evaluating a similarity with a candidate item based on the set-inclusion criteria and basing a recommendation regarding the candidate item on the similarity.

Scoring candidates for set recommendation problems

Growing community of inventors

San Jose, CA, United States of America

Rasmus Munk Larsen