Inventor

SAINATH TARA N

US143 patents

⚠️ This page may combine multiple inventors who share the name “SAINATH TARA N”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

SAINATH TARA N

GOOGLE LLC

Automatic speech recognition using multi-dimensional models

Enhanced multi-channel acoustic models

Complex linear projection for acoustic modeling

Multi-dialect and multilingual speech recognition

Minimum word error rate training for attention-based sequence-to-sequence models

Contextual biasing for speech recognition

Phoneme-based contextualization for cross-lingual speech recognition in end-to-end models

Contextual biasing for speech recognition using grapheme and phoneme data

Minimum word error rate training for attention-based sequence-to-sequence models

Multichannel speech recognition using neural networks

Multi-dialect and multilingual speech recognition

Speech recognition with sequence-to-sequence models

Adaptive audio enhancement for multichannel speech recognition

Processing audio waveforms

Convolutional neural networks

Large-scale language model data selection for rare-word speech recognition

Deliberation model-based two-pass end-to-end speech recognition

Contextual biasing for speech recognition

Convolutional, long short-term memory, fully connected deep neural networks

Adaptive audio enhancement for multichannel speech recognition

Convolutional, long short-term memory, fully connected deep neural networks

Compressed recurrent neural network models

Multichannel raw-waveform neural networks

IBM

Managing vehicles on a road network

Managing vehicles on a road network

Reducing cognitive demand on a vehicle operator by generating passenger stimulus

Adjusting presentation of content on a display

Systems and methods for accelerating hessian-free optimization for deep neural networks by implicit preconditioning and sampling

Automatically generating email subject lines

Improving voice communication over a network

Remote object sensing in video

Protecting network entity data while preserving network properties

System and method for using biometrics to predict and select music preferences

Tagging scanned data with emotional tags, predicting emotional reactions of users to data, and updating historical user emotional reactions to data

Failure impact manager

GOOGLE INC

Processing multi-channel audio waveforms

Low-rank hidden input layer for speech recognition neural network

Adaptive audio enhancement for multichannel speech recognition

User specified keyword spotting using neural network feature extractor

User specified keyword spotting using long short term memory neural network feature extractor

KANEVSKY DIMITRI

System and method for in-private browsing

Sparse representation features for speech recognition

Phonetic features for speech recognition

BASSON SARA H

Intelligent dialogue amongst competitive user applications

Text processing using natural language understanding

System for preventing handheld device use while operating a vehicle

NUANCE COMMUNICATIONS INC

Text processing using natural language understanding

Hybrid pre-training of deep belief networks

CASKEY SASHA P

Providing multi-lingual searching of mono-lingual content

GUPTA SACHIN

Multi-dimensional heating and cooling system