Inventor
Tara N. Sainath
Inventor
New York, NY, US
Overview
Patents· 91 as inventor
Public. Assignee & technology derived from the patent record. Each patent links to its LexDana page.
G06Fprimary
G06FElectric digital data processing22%
G10LSpeech analysis techniques or speech synthesis22%
G06NComputing arrangements based on specific computational models9%
H04LTransmission of digital information, e.g. telegraphic communication7%
G06QInformation and communication technology [ict] specially adapted for administrative, commercial, financial, managerial or supervisory purposes5%
G06VImage or video recognition or understanding4%
G06KGraphical data reading3%
G05BControl or regulating systems in general2%
G09BEducational or demonstration appliances2%
G08GTraffic control systems2%
PatentTitleYear
11,107,463Minimum word error rate training for attention-based sequence-to-sequence models202111,062,725Multichannel speech recognition using neural networks202110,930,270Processing audio waveforms202110,783,900Convolutional, long short-term memory, fully connected deep neural networks202010,762,894Convolutional neural networks202010,714,078Linear transformation for speech recognition modeling202010,515,626Adaptive audio enhancement for multichannel speech recognition201910,515,307Compressed recurrent neural network models201910,490,198Device-specific multi-channel data compression neural network201910,403,269Processing audio waveforms201910,360,901Learning front-end speech recognition parameters within neural network training201910,339,921Multichannel raw-waveform neural networks201910,229,700Voice activity detection201910,224,058Enhanced multi-channel acoustic models201910,180,974System and method for generating content corresponding to an event201910,140,980Complex linear projection for acoustic modeling201810,140,371Providing multi-lingual searching of mono-lingual content201810,078,636Providing a human-sense perceivable representation of an aspect of an event201810,068,492Controlled resources based on good behavior201810,056,075Systems and methods for accelerating hessian-free optimization for deep neural networks by implicit preconditioning and sampling201810,007,887Failure impact manager20189,984,683Automatic speech recognition using multi-dimensional models20189,984,104Indexing content and source code of a software application20189,886,949'Adaptive audio enhancement for multichannel speech recognition'20189,875,747Device specific multi-channel data compression20189,792,367'Providing multi-lingual searching of mono-lingual content'20179,785,862'System and method for feeling surface roughness through microscopic data'20179,779,080'Text auto-correction via N-grams'20179,771,082'Reducing cognitive demand on a vehicle operator by generating passenger stimulus'20179,754,584'User specified keyword spotting using neural network feature extractor'20179,697,826'Processing multi-channel audio waveforms'20179,691,361'Adjusting presentation of content on a display'20179,646,634'Low-rank hidden input layer for speech recognition neural network'20179,640,186'Deep scattering spectrum in acoustic modeling for speech recognition'20179,601,109'Systems and methods for accelerating hessian-free optimization for deep neural networks by implicit preconditioning and sampling'20179,507,326'System and method for using biometrics to predict and select music preferences'20169,508,340'User specified keyword spotting using long short term memory neural network feature extractor'20169,414,779'Electronic communication warning and modification'20169,418,158'Providing multi-lingual searching of mono-lingual content'20169,418,334'Hybrid pre-training of deep belief networks'20169,402,576'Electronic communication warning and modification'20169,389,595'System and method for using biometrics to predict and select music preferences'20169,378,464'Discriminative learning via hierarchical transformations'20169,373,324'Applying speaker adaption techniques to correlated features'20169,367,576'Indexing of significant media granulars'20169,356,889'Automatically generating email subject lines'20169,331,965'Automatically generating email subject lines'20169,313,150'Generating intentional interruptions during gatherings'20169,306,243'Optimizing battery usage'20169,305,038'Indexing of significant media granulars'20169,299,268'Tagging scanned data with emotional tags, predicting emotional reactions of users to data, and updating historical user emotional reactions to data'20169,293,133'Improving voice communication over a network'20169,291,474'System and method for providing global positioning system (GPS) feedback to a user'20169,286,338'Indexing content and source code of a software application'20169,286,889'Improving voice communication over a network'20169,275,424'Controlled resources based on good behavior'20169,275,546'System and method for minimizing the time to park a vehicle'20169,269,932'Power source encapsulation'20169,270,940'Remote object sensing in video'20169,262,724'Low-rank matrix factorization for deep belief network training with high-dimensional output targets'20169,263,714'Power source encapsulation'20169,244,441'System and method for using biometrics to predict and select music preferences'20169,235,198'System and method for using biometrics to predict and select music preferences'20169,232,016'Undoing sent communications'20169,232,019'Undoing sent communications'20169,219,702'System and method for generating intentional interruptions during gatherings'20159,195,650'Translating between spoken and written language'20159,177,215'Sparse representation for dynamic sensor networks'20159,152,154'Multi-dimensional heating and cooling system'20159,122,548'Clipboard for processing received data content'20159,077,813'Masking mobile message content'20159,008,952'Managing vehicles on a road network'20158,924,210'Text processing using natural language understanding'20148,918,308'Providing multi-lingual searching of mono-lingual content'20148,914,225'Managing vehicles on a road network'20148,902,792'Forming a convex polygon of mobile sensors'20148,898,447'Computer device separable into multiple sub-computers'20148,892,864'Method for separating a dividable computer device into multiple sub-computers with predetermined features and functionality loaded before the separation upon user\'s selection'20148,867,707'Automatically updating meeting information'20148,856,004'Text processing using natural language understanding'20148,856,002'Distance metrics for universal pattern processing tasks'20148,825,533'Intelligent dialogue amongst competitive user applications'20148,769,009'Virtual communication techniques'20148,566,270'Sparse representations for text classification'20138,547,214'System for preventing handheld device use while operating a vehicle'20138,527,566'Directional optimization via EBW'20138,484,024'Phonetic features for speech recognition'20138,484,023'Sparse representation features for speech recognition'20138,464,350'System and method for in-private browsing'20138,346,774'Protecting network entity data while preserving network properties'20138,326,668'Managing encounters with persons'2012Career & activity timeline
Assembled from patent assignee/location history and declaration credentials.
2012–2019Patents assigned to IBM (62)· patent assignee history
2012–2017Patents assigned to KANEVSKY DIMITRI (26)· patent assignee history
2012–2017Patents assigned to SAINATH TARA N (26)· patent assignee history
2012–2018Based in New York, NY, US· patent location history
2013–2019Based in Burlington, MA, US· patent location history
2013Based in Ossining, NY, US· patent location history
2014–2017Patents assigned to CASKEY SASHA P (9)· patent assignee history
2014Based in Yorktown Heights, NY, US· patent location history
2016–2021Based in Jersey City, NJ, US· patent location history
2018–2021Patents assigned to GOOGLE LLC (16)· patent assignee history
Inventor activity
Assignee and location history derived from the patent record.
Assignees
| Assignee | Patents | Span |
|---|---|---|
| IBM | 62 | 2012–2019 |
| KANEVSKY DIMITRI | 26 | 2012–2017 |
| SAINATH TARA Nself | 26 | 2012–2017 |
| GOOGLE LLC | 16 | 2018–2021 |
| CASKEY SASHA P | 9 | 2014–2017 |
Locations
| Location | Patents | Span |
|---|---|---|
| New York, New York, US | 51 | 2012–2018 |
| Jersey City, New Jersey, US | 31 | 2016–2021 |
| Burlington, Massachusetts, US | 6 | 2013–2019 |
| Yorktown Heights, New York, US | 2 | 2014 |
| Ossining, New York, US | 1 | 2013 |