Snap Inc.

Xin Chen

Jaewook Chung

Wei Chu

Zehao Xue

Chen Cao

Ning Xu

Kirk Ouimet

Xing Mei

Wisam Dakka

Emre Yamangil

Yanping Chen

Yu Hu

Jiwoon Jeon

Theresa Barton

Keyang Zhang

Lucas Ou-Yang

Jinhua Jiang

Yurii Monastyrshyn

Fedir Poliakov

Leonardo Ribas Machado Das Neves

William Brendel

Luis Carlos Dos Santos Marujo

Shubham Vij

Venkata Satya Pradeep Karuturi

Francesco Barbieri

The present invention relates to a joint automatic audio visual driven facial animation system that in some example embodiments includes a full scale state of the art Large Vocabulary Continuous Speech Recognition (LVCSR) with a strong language model for speech recognition and obtained phoneme alignment from the word lattice.

Joint audio-video facial animation system

Disclosed are methods and systems for ranking content. In one aspect, a method of ranking content for display includes identifying, via hardware processing circuitry, interactions by a single account with content pairs, each of the content in the content pairs included in a plurality of content, aggregating, via the hardware processing circuitry, the identified interactions across a plurality of accounts, associating, via the hardware processing circuitry, probabilities with each content in the plurality of content based on the aggregated interactions, ranking, via the hardware processing circuitry, the plurality of content based on the associated probabilities; and selecting, via the hardware processing circuitry, content ranked above a threshold for display.

Ranking content for display

An audio control system can control interactions with an application or device using keywords spoken by a user of the device. The audio control system can use machine learning models (e.g., a neural network model) trained to recognize one or more keywords. Which machine learning model is activated can depend on the active location in the application or device. Responsive to detecting keywords, different actions are performed by the device, such as navigation to a pre-specified area of the application.

Device control using audio data

Symbol prediction can be implemented using a multi-task system trained for different tasks. The tasks may include a single symbol prediction, symbol category prediction, and symbol subcategory prediction. Categories of symbols can be generated by clustering sets of training data using a clustering scheme.

Pictorial symbol prediction

Systems and methods provide for analyzing at least one feature of each segment of a plurality of segments of the audio stream to generate for each segment a prediction value indicating whether there is music in the segment, generating a probability value that there is music in the audio stream based on aggregating the prediction values of the plurality of segments, and causing the audio stream to be identified based on determining that the probability value that there is music in the audio stream meets a predetermined threshold.

Generating a probability of music

Systems and Methods describe capturing a plurality of segments of an audio stream, analyzing at least one feature of each segment of the plurality of segments of the audio stream to generate for each segment a prediction value indicating whether there is music in the segment, generating a probability value that there is music in the audio stream based on aggregating the prediction values of the plurality of segments, determining that the probability value that there is music in the audio stream meets a predetermined threshold, and causing the audio stream to be identified based on determining that the probability value that there is music in the audio stream meets a predetermined threshold.

Systems and methods provide for capturing a plurality of segments of an audio stream and, for each segment of the plurality of segments of the audio stream: performing feature extraction on an audio signal of the segment using a feature extraction machine learning model that analyzes the audio signal to generate a feature vector for the segment and generating a prediction value for the segment for whether there is music in the segment using the extracted feature vector and a music detector machine learning model. The systems and methods further provide for generating a probability value that there is music in the audio stream based on the prediction value for each of the plurality of segments and causing the audio stream to be identified based on determining that the probability value that there is music in the audio stream meets a predetermined threshold.

Generating a probability of music using machine learning technology

Growing community of inventors

Torrance, CA, United States of America

Xin Chen