Inventor · disambiguated record
Samuel Thomas
Also filed as: SAMOUR CARLOS M · THOMAS SAMUEL · THOMAS SAMUEL J · THOMAS SAMUEL JONATHAN
43 granted patents·4 pending applications·290 citations·filing 1978–2023
97Inventor score
Top patents by PatentIndex Score
47 records- 0196US9584758B1Combining installed audio-visual sensors with ad-hoc mobile audio-visual sensors for smart meeting roomsIBM·Filed 2015·Granted Feb 28, 2017·18 cites·20 claims
- 0292US12136414B2Integrating dialog history into end-to-end spoken language understanding systemsIBM·Filed 2021·Granted Nov 5, 2024·3 cites·20 claims
- 0392US12046236B2Training end-to-end spoken language understanding systems with unordered entitiesIBM·Filed 2021·Granted Jul 23, 2024·3 cites·17 claims
- 0492US9697823B1Acoustic model trainingIBM·Filed 2016·Granted Jul 4, 2017·12 cites·11 claims
- 0591US11929062B2End-to-end spoken language understanding without full transcriptsIBM·Filed 2020·Granted Mar 12, 2024·3 cites·20 claims
- 0691US11908454B2Integrating text inputs for training and adapting neural network transducer ASR modelsIBM·Filed 2021·Granted Feb 20, 2024·2 cites·20 claims
- 0790US9697835B1Acoustic model trainingIBM·Filed 2016·Granted Jul 4, 2017·9 cites·16 claims
- 0889US10546575B2Using recurrent neural network for partitioning of audio data into segments that each correspond to a speech feature cluster identifierIBM·Filed 2016·Granted Jan 28, 2020·9 cites·20 claims
- 0989US10380484B2Annealed dropout training of neural networksIBM·Filed 2015·Granted Aug 13, 2019·10 cites·12 claims
- 1089US4213486ACoated pipe and process for making sameKENDALL & CO·Filed 1978·Granted Jul 22, 1980·61 cites·16 claims
- 1188US10249292B2Using long short-term memory recurrent neural network for speaker diarization segmentationIBM·Filed 2016·Granted Apr 2, 2019·9 cites·20 claims
- 1282US11019306B2Combining installed audio-visual sensors with ad-hoc mobile audio-visual sensors for smart meeting roomsIBM·Filed 2019·Granted May 25, 2021·2 cites·15 claims
- 1382US4523141APipe coatingKENDALL & CO·Filed 1982·Granted Jun 11, 1985·40 cites·1 claims
- 1480US11250872B2Using closed captions as parallel training data for customization of closed captioning systemsIBM·Filed 2019·Granted Feb 15, 2022·2 cites·20 claims
- 1578US11907845B2Training teacher machine learning models using lossless and lossy branchesIBM·Filed 2020·Granted Feb 20, 2024·1 cites·18 claims
- 1678US10347270B2Denoising a signalIBM·Filed 2016·Granted Jul 9, 2019·3 cites·7 claims
- 1777US11410029B2Soft label generation for knowledge distillationIBM·Filed 2018·Granted Aug 9, 2022·3 cites·19 claims
- 1874US9912909B2Combining installed audio-visual sensors with ad-hoc mobile audio-visual sensors for smart meeting roomsIBM·Filed 2016·Granted Mar 6, 2018·1 cites·19 claims
- 1972US11183194B2Detecting and recovering out-of-vocabulary words in voice-to-text transcription systemsIBM·Filed 2019·Granted Nov 23, 2021·2 cites·13 claims
- 2072US10373054B2Annealed dropout training of neural networksIBM·Filed 2015·Granted Aug 6, 2019·2 cites·19 claims
- 2172US4268334APipe wrapping adhesive tape methodKENDALL & CO·Filed 1979·Granted May 19, 1981·21 cites·7 claims
- 2270US10629221B2Denoising a signalIBM·Filed 2019·Granted Apr 21, 2020·1 cites·20 claims
- 2369US10902843B2Using recurrent neural network for partitioning of audio data into segments that each correspond to a speech feature cluster identifierIBM·Filed 2019·Granted Jan 26, 2021·1 cites·20 claims
- 2469US10657980B2Denoising a signalIBM·Filed 2017·Granted May 19, 2020·1 cites·11 claims
- 2567US11416741B2Teacher and student learning for constructing mixed-domain modelIBM·Filed 2018·Granted Aug 16, 2022·1 cites·24 claims
- 2667US10096315B2Acoustic model trainingIBM·Filed 2017·Granted Oct 9, 2018·1 cites·20 claims
- 2767US4589275APipe coatingKENDALL & CO·Filed 1984·Granted May 20, 1986·22 cites·3 claims
- 2865US10984814B2Denoising a signalIBM·Filed 2020·Granted Apr 20, 2021·0 cites·20 claims
- 2965US7426631B2Methods and systems for storing branch information in an address table of a processorIBM·Filed 2005·Granted Sep 16, 2008·2 cites·8 claims
- 3063US5352292AApparatus for coating exterior weld joints of a pipeKENDALL & CO·Filed 1993·Granted Oct 4, 1994·33 cites·2 claims
- 3162US12387717B2Multi-speaker data augmentation for improved end-to-end automatic speech recognitionIBM·Filed 2023·Granted Aug 12, 2025·0 cites·20 claims
- 3261US9959887B2Multi-pass speech activity detection strategy to improve automatic speech recognitionIBM·Filed 2016·Granted May 1, 2018·1 cites·19 claims
- 3360US12444405B2Textual knowledge transfer for improved speech recognition and understandingIBM·Filed 2023·Granted Oct 14, 2025·0 cites·20 claims
- 3457US12119008B2End-to-end integration of dialog history for spoken language understandingIBM·Filed 2022·Granted Oct 15, 2024·0 cites·20 claims
- 3557US10230922B2Combining installed audio-visual sensors with ad-hoc mobile audio-visual sensors for smart meeting roomsIBM·Filed 2017·Granted Mar 12, 2019·0 cites·15 claims
- 3656US11900922B2Multilingual intent recognitionIBM·Filed 2020·Granted Feb 13, 2024·0 cites·18 claims
- 3756US11587551B2Leveraging unpaired text data for training end-to-end spoken language understanding systemsIBM·Filed 2020·Granted Feb 21, 2023·0 cites·8 claims
- 3855US2023153601A1Global neural transducer models leveraging sub-task networksIBM·Filed 2021·Application pending·0 cites
- 3954US7984280B2Storing branch information in an address table of a processorIBM·Filed 2008·Granted Jul 19, 2011·0 cites·6 claims
- 4051US11610108B2Training of student neural network with switched teacher neural networksIBM·Filed 2018·Granted Mar 21, 2023·0 cites·14 claims
- 4151US11568858B2Transliteration based data augmentation for training multilingual ASR acoustic models in low resource settingsIBM·Filed 2020·Granted Jan 31, 2023·0 cites·18 claims
- 4250US8943301B2Storing branch information in an address table of a processorKONIGSBURG BRIAN R·Filed 2011·Granted Jan 27, 2015·0 cites·19 claims
- 4349US11741355B2Training of student neural network with teacher neural networksIBM·Filed 2018·Granted Aug 29, 2023·0 cites·20 claims
- 4446US2022414448A1Cross-lingual knowledge transfer learningIBM·Filed 2021·Application pending·0 cites
- 4545US2022319494A1End to end spoken language understanding modelIBM·Filed 2021·Application pending·0 cites
- 4644US4780158ANovel method for heating cylindrical articles and apparatus useful thereforKENDALL & CO·Filed 1987·Granted Oct 25, 1988·11 cites·8 claims
- 4738US2017154264A1Autonomous collaboration agent for meetingsIBM·Filed 2015·Application pending·0 cites
Identity basis: PatentsView inventor disambiguation (2025Q4-odp release). How scoring works →