Inventor · disambiguated record
Dong Yu
Also filed as: YU DONG · YU DONG W · YU DONG WON
61 granted patents·12 pending applications·215 citations·filing 1984–2025
98Inventor score
Files withTencent America LLC66MICROSOFT TECHNOLOGY LICENSING LLC23SIXTY INNOVATIONS LLC1KOREA ADVANCED INST SCI & TECH1MICROSOFT CORP1
Top patents by PatentIndex Score
73 records- 0195US9818431B2Multi-speaker speech separationMICROSOFT TECHNOLOGY LICENSING LLC·Filed 2015·Granted Nov 14, 2017·34 cites·21 claims
- 0294US9390712B2Mixed speech recognitionMICROSOFT CORP·Filed 2014·Granted Jul 12, 2016·27 cites·20 claims
- 0393US12094481B2ADL-UFE: all deep learning unified front-end systemTencent America LLC·Filed 2021·Granted Sep 17, 2024·5 cites·20 claims
- 0493US11902757B2Techniques for unified acoustic echo suppression using a recurrent neural networkTencent America LLC·Filed 2022·Granted Feb 13, 2024·2 cites·20 claims
- 0591US11302301B2Learnable speed control for speech synthesisTencent America LLC·Filed 2020·Granted Apr 12, 2022·3 cites·16 claims
- 0691US11183168B2Singing voice conversionTencent America LLC·Filed 2020·Granted Nov 23, 2021·3 cites·16 claims
- 0788US9824684B2Prediction-based sequence recognitionMICROSOFT TECHNOLOGY LICENSING LLC·Filed 2014·Granted Nov 21, 2017·16 cites·13 claims
- 0887US11380307B2All deep learning minimum variance distortionless response beamformer for speech separation and enhancementTencent America LLC·Filed 2020·Granted Jul 5, 2022·2 cites·17 claims
- 0985US11410652B2Multi-look enhancement modeling and application for keyword spottingTencent America LLC·Filed 2020·Granted Aug 9, 2022·2 cites·14 claims
- 1084US11688412B2Multi-modal framework for multi-channel target speech separationTencent America LLC·Filed 2020·Granted Jun 27, 2023·2 cites·20 claims
- 1184US11423906B2Multi-tap minimum variance distortionless response beamformer with neural networks for target speech separationTencent America LLC·Filed 2020·Granted Aug 23, 2022·2 cites·14 claims
- 1283US11004443B2Multistage curriculum training framework for acoustic-to-word speech recognitionTencent America LLC·Filed 2018·Granted May 11, 2021·6 cites·18 claims
- 1382US11776556B2Unified deep neural network model for acoustic echo cancellation and residual echo suppressionTencent America LLC·Filed 2021·Granted Oct 3, 2023·1 cites·17 claims
- 1480US11170167B2Automatic lexical sememe prediction system using lexical dictionariesTencent America LLC·Filed 2019·Granted Nov 9, 2021·2 cites·16 claims
- 1580US4675434AProcess for producing phthalic acid estersKOREA ADVANCED INST SCI & TECH·Filed 1984·Granted Jun 23, 1987·25 cites·11 claims
- 1679US4908222AMicrowave brewing apparatus and methodYU DONG·Filed 1989·Granted Mar 13, 1990·56 cites·10 claims
- 1777US2025356871A1Kalmannet: a learnable kalman filter for acoustic echo cancellationTencent America LLC·Filed 2025·Application pending·0 cites
- 1876US12293769B2Neural-echo: an unified deep neural network model for acoustic echo cancellation and residual echo suppressionTencent America LLC·Filed 2023·Granted May 6, 2025·0 cites·20 claims
- 1975US12321188B2Neuralecho: a self-attentive recurrent neural network for unified acoustic echo suppression, speaker aware speech enhancement and automatic gain controlTencent America LLC·Filed 2023·Granted Jun 3, 2025·0 cites·20 claims
- 2074US12293154B2Extractive method for speaker identification in texts with self-trainingTencent America LLC·Filed 2024·Granted May 6, 2025·0 cites·15 claims
- 2174US11037547B2Token-wise training for attention based end-to-end speech recognitionTencent America LLC·Filed 2019·Granted Jun 15, 2021·2 cites·17 claims
- 2273US11803618B2N-best softmax smoothing for minimum bayes risk training of attention based sequence-to-sequence modelsTencent America LLC·Filed 2022·Granted Oct 31, 2023·0 cites·20 claims
- 2371US12468889B2Multi-task self-training for character gender identificationTencent America LLC·Filed 2023·Granted Nov 11, 2025·0 cites·20 claims
- 2471US12468896B2Method and apparatus for zero-shot natural language processing using visual imaginationTencent America LLC·Filed 2022·Granted Nov 11, 2025·0 cites·20 claims
- 2570US12507005B2Elevating acoustic howling suppression via recursive neural network trainingTencent America LLC·Filed 2023·Granted Dec 23, 2025·0 cites·20 claims
- 2670US12406683B2Kalmannet: a learnable Kalman filter for acoustic echo cancellationTencent America LLC·Filed 2023·Granted Sep 2, 2025·0 cites·20 claims
- 2770US12308019B2Learning singing from speechTencent America LLC·Filed 2022·Granted May 20, 2025·0 cites·16 claims
- 2870US11011154B2Enhancing hybrid self-attention structure with relative-position-aware bias for speech synthesisTencent America LLC·Filed 2019·Granted May 18, 2021·2 cites·17 claims
- 2969US2025308511A1Robust disentangled variational speech representation learning for zero-shot voice conversionTencent America LLC·Filed 2025·Application pending·0 cites
- 3068US12308039B2Multi-band synchronized neural vocoderTencent America LLC·Filed 2022·Granted May 20, 2025·0 cites·20 claims
- 3168US11128435B2Distributed and collaborative analytics of encrypted data using deep polynomial networksTencent America LLC·Filed 2019·Granted Sep 21, 2021·1 cites·17 claims
- 3267US11295751B2Multi-band synchronized neural vocoderTencent America LLC·Filed 2019·Granted Apr 5, 2022·1 cites·11 claims
- 3367US11151979B2Duration informed attention network (DURIAN) for audio-visual synthesisTencent America LLC·Filed 2019·Granted Oct 19, 2021·1 cites·18 claims
- 3466US12154566B2Multi-look enhancement modeling and application for keyword spottingTencent America LLC·Filed 2022·Granted Nov 26, 2024·0 cites·20 claims
- 3566US11610060B2Automatic lexical sememe prediction system using lexical dictionariesTencent America LLC·Filed 2021·Granted Mar 21, 2023·0 cites·18 claims
- 3666US11257481B2Multi-task training architecture and strategy for attention-based speech recognition systemTencent America LLC·Filed 2018·Granted Feb 22, 2022·1 cites·20 claims
- 3765US11682379B2Learnable speed control of speech synthesisTencent America LLC·Filed 2022·Granted Jun 20, 2023·0 cites·20 claims
- 3865US11670283B2Duration informed attention network (DURIAN) for audio-visual synthesisTencent America LLC·Filed 2021·Granted Jun 6, 2023·0 cites·18 claims
- 3965US2025190469A1Instance-level adaptive propulsion of external knowledge (iapek)Tencent America LLC·Filed 2025·Application pending·0 cites
- 4064US12456457B2All deep learning minimum variance distortionless response beamformer for speech separation and enhancementTencent America LLC·Filed 2022·Granted Oct 28, 2025·0 cites·20 claims
- 4163US12475914B2Conditional factorization for jointly modeling code-switched and monolingual ASRTencent America LLC·Filed 2022·Granted Nov 18, 2025·0 cites·17 claims
- 4263US12373643B2Efficient zero shot event extraction with context definitionTencent America LLC·Filed 2022·Granted Jul 29, 2025·0 cites·18 claims
- 4363US11636848B2Token-wise training for attention based end-to-end speech recognitionTencent America LLC·Filed 2021·Granted Apr 25, 2023·0 cites·17 claims
- 4463US2025238693A1Moe model from block sparse computation's point of viewTencent America LLC·Filed 2024·Application pending·0 cites
- 4562US12265564B2Instance-level adaptive propulsion of external knowledge (IAPEK)Tencent America LLC·Filed 2022·Granted Apr 1, 2025·0 cites·15 claims
- 4662US12001795B2Extractive method for speaker identification in texts with self-trainingTencent America LLC·Filed 2021·Granted Jun 4, 2024·0 cites·15 claims
- 4762US11721318B2Singing voice conversionTencent America LLC·Filed 2021·Granted Aug 8, 2023·0 cites·20 claims
- 4862US11430431B2Learning singing from speechTencent America LLC·Filed 2020·Granted Aug 30, 2022·0 cites·20 claims
- 4961US12354594B2Techniques for disentangled variational speech representation learning for zero-shot voice conversionTencent America LLC·Filed 2022·Granted Jul 8, 2025·0 cites·20 claims
- 5061US12347218B2Salience-aware cross-attention for abstractive summarizationTencent America LLC·Filed 2022·Granted Jul 1, 2025·0 cites·20 claims
Showing the top 50 of 73 patent records by PatentIndex Score.
Join the waitlist — get patent alerts
Get an alert when Dong Yu files or is granted a new patent.
We store only your email — no account needed. See our privacy policy.
Identity basis: PatentsView inventor disambiguation (2025Q4-odp release). How scoring works →