Inventor · Kirkland, WA, US

Nikko Strom

45Patents
18h-index
97Co-inventors
84Inventor score

Filing activity: Mar 22, 2001 → Apr 28, 2023

Most-cited inventions

PatentTitleAreaCited byStatus
US7630900B1 Method and system for selecting grammars based on geographic information associated with a caller Electricity 372 Active
US6985862B2 Histogram grammar weighting and error corrective training of grammar weights Physics 254 Expired
US6873952B1 Coarticulated concatenated speech Physics 246 Expired
US7644000B1 Adding audio effects to spoken utterance Physics 234 Active
US7269557B1 Coarticulated concatenated speech Physics 230 Expired
US8983844B1 Transmission of noise parameters for improving automatic speech recognition Physics 202 Active
US9818407B1 Distributed endpointing for speech recognition Physics 162 Active
US8898064B1 Identifying candidate passwords from captured audio Physics 137 Active
US10847137B1 Trigger word detection using neural network waveform processing Physics 67 Active
US8972243B1 Parse information encoding in a finite state transducer Physics 66 Active
US10109273B1 Efficient generation of personalized spoken language understanding models Physics 47 Active
US10152676B1 Distributed training of models using stochastic gradient descent Physics 42 Active
US8036902B1 Audio human verification Physics 36 Active
US10032463B1 Speech processing with learned representation of user interaction history Physics 34 Active
US10297250B1 Asynchronous transfer of audio data Physics 31 Active
US11087739B1 On-device learning in a hybrid speech processing system Physics 20 Active
US10354184B1 Joint modeling of user behavior Physics 19 Active
US9653093B1 Generative modeling of speech using neural networks Physics 19 Active
US10726830B1 Deep multi-channel acoustic modeling Physics 18 Active
US9286897B2 Speech recognizer with multi-directional decoding Electricity 18 Active
US11200885B1 Goal-oriented dialog system Physics 15 Active
US9600764B1 Markov-based sequence tagging using neural networks Physics 12 Active
US11574628B1 Deep multi-channel acoustic modeling using multiple microphone array geometries Electricity 10 Active
US8386251B2 Progressive application of knowledge sources in multistage speech recognition Physics 10 Active
US10510358B1 Resolution enhancement of speech signals for speech synthesis Physics 8 Active

Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.