Inventor

PANG RUOMING

US44 patents

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

PANG RUOMING

GOOGLE LLC

Neural architecture search with factorized hierarchical search space

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

Attention-based joint acoustic and text on-device end-to-end model

Synthesis of speech from text in a voice of a target speaker using neural networks

Synthesizing speech from text using neural networks

Efficient streaming non-recurrent on-device end-to-end model

Unsupervised learning of disentangled speech content and style representation

Asynchronous distributed data flow for machine learning workloads

Neural architecture search with factorized hierarchical search space

Synthesis of speech from text in a voice of a target speaker using neural networks

Synthesizing speech from text using neural networks

Transducer-based streaming deliberation for cascaded encoders

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

Neural architecture search with factorized hierarchical search space

Synthesis of speech from text in a voice of a target speaker using neural networks

Emitting word timings with end-to-end models

Efficient streaming non-recurrent on-device end-to-end model

Emitting word timings with end-to-end models

Convolution-augmented transformer models

Convolution-augmented transformer models

Emitting word timings with end-to-end models

Asynchronous distributed data flow for machine learning workloads

Hardware-optimized neural architecture search

Systems and methods for training dual-mode machine-learned speech recognition models

Fast emit low-latency streaming ASR with sequence-level emission regularization utilizing forward and backward probabilities between nodes of an alignment lattice

Two-pass end to end speech recognition

Cascaded encoders for simplified streaming and non-streaming ASR

Two-pass end to end speech recognition

Optimizing inference performance for conformer

Streaming automatic speech recognition with non-streaming model distillation

Using context information with end-to-end models for speech recognition

Language agnostic multilingual end-to-end streaming on-device ASR system

Single-stage model training for neural architecture search

MICROSOFT CORP

Integrity ordainment and ascertainment of computer-executable instructions with consideration for execution context

Integrity ordainment and ascertainment of computer-executable instructions with consideration for execution context

Integrity ordainment and ascertainment of computer-executable instructions

Integrity ordainment and ascertainment of computer-executable instructions with consideration for execution context

GABBER ERAN

Identifying transient portions of web pages

Identifying transient paths within websites

GOOGLE INC

Systems and methods for live migration of a data store to encrypted form

Efficient indexing and searching of access control listed documents

FLASTER MICHAEL

Efficient document clustering

SANDLER MARK M

Monitoring of negative feedback systems

KORN JEFFREY

Efficient indexing and searching of access control listed documents