Suki Ai, Inc.

Ayasdi Ai LLC

Symphonyai Sensa LLC

Ajithkumar Warrier

Vamsi Reddy Chagari

Ryan Hsu

Jian Feng

Mona Abdul-Azeez Sheikh

Sudhakar Jonnalagadda

Jennifer Kloke

Gaurav Trivedi

Valerii Garnaga

Priyank Raj

Raghu P Chalasani

Systems and methods to recognize pronunciation variations of individual terms are disclosed. Exemplary implementations may: effectuate presentation of a diction interface; obtain audio information that represents sounds captured by the client computing platform; perform speech recognition on the audio information to generate a first term spelling and first term recognition spellings; determine a first ground truth spelling of the first term; upon determination that a first user-specific lexicon does not include the first ground truth spelling, add the first ground truth spelling to first user-specific lexicon; upon the first term recognition spellings matching the first ground truth spelling, correlate the utterance of the first term and the first ground truth spellings; upon the first term recognition spellings not matching the first ground truth spelling, correlate the first term recognition spellings and the first ground truth spelling, and correlate the first ground truth spelling and the utterance of the first term.

Systems and methods to recognize pronunciation variations of individual terms

Systems and methods to improve automatic speech recognition of audio data are disclosed. Exemplary implementations may: process a first noisy audio data from a set of the noisy audio data, with a generator, to generate first refined audio data; process the first refined audio data and a first comprehensible audio data from a set of the comprehensible audio data, individually with a discriminator, to determine individual decision values; determine a generator update and a discriminator update based on the individual decision values; apply the generator update to the generator such that the generator is modified to improve generation of refined audio data or is not modified; and apply the discriminator update to the discriminator such that the discriminator is modified to improve subsequent determinations of the decision values or is not modified.

Systems and methods to improve automatic speech recognition of audio data

A method comprises receiving a network of a plurality of nodes and a plurality of edges, each of the nodes of the plurality of nodes comprising members representative of at least one subset of initial data points, selecting a subset of the data points based on each node of the plurality of nodes, for each selected data point of the set of selected data points, determining a predetermined number of other data points that are closest in distance to that particular selected data point, grouping the selected data points into a plurality of groups based, at least in part, on the predetermined number of other data points of the set of selected data points that are closest in distance, each group of the plurality of groups including a different subset of data points, and providing a list of selected data points and the plurality of groups.

Systems and methods for topological data analysis using nearest neighbors

Growing community of inventors

Fremont, CA, United States of America

Ajithkumar Warrier