Inventor
Michael Levit
Inventor
San Jose, CA, US
Overview
Patents· 21 as inventor
Public. Assignee & technology derived from the patent record. Each patent links to its LexDana page.
G10Lprimary
G10LSpeech analysis techniques or speech synthesis51%
G06FElectric digital data processing35%
G06NComputing arrangements based on specific computational models5%
G06QInformation and communication technology [ict] specially adapted for administrative, commercial, financial, managerial or supervisory purposes3%
G16BBioinformatics, i.e. information and communication technology [ict] specially adapted for genetic or protein-related data processing in computational molecular biology3%
G16CComputational chemistry3%
PatentTitleYear
11,120,788Organizational-based language model generation202110,497,367Flexible schema for language model customization201910,192,545Language modeling based on spoken and unspeakable corpuses20199,972,311Language model optimization for in-domain application20189,922,654'Incremental utterance decoder combination for efficient and accurate decoding'20189,761,220'Language modeling based on spoken and unspeakable corpuses'20179,734,826'Token-level interpolation for class-based language models'20179,672,202'Context-aware re-formating of an input'20179,552,817'Incremental utterance decoder combination for efficient and accurate decoding'20179,542,931'Leveraging interaction context to improve recognition confidence scores'20179,529,794'Flexible schema for language model customization'20169,460,081'Transcription correction using multi-token structures'20169,384,188'Transcription correction using multi-token structures'20169,299,342'User query history expansion for improving language model adaptation'20169,129,606'User query history expansion for improving language model adaptation'20159,053,087'Automatic semantic evaluation of speech recognition results'20158,930,179'Recognition using re-recognition and statistical classification'20158,886,532'Leveraging interaction context to improve recognition confidence scores'20148,260,817'Semantic matching using predicate-argument structure'20128,180,641'Sequential speech recognition with two unequal ASR systems'20127,890,539'Semantic matching using predicate-argument structure'2011Career & activity timeline
Assembled from patent assignee/location history and declaration credentials.
2011–2012Based in Berkeley, CA, US· patent location history
2012–2015Patents assigned to LEVIT MICHAEL (6)· patent assignee history
2012–2018Patents assigned to MICROSOFT CORP (5)· patent assignee history
2012–2015Patents assigned to BUNTSCHUH BRUCE MELVIN (4)· patent assignee history
2012–2015Patents assigned to CHANG SHUANGYU (4)· patent assignee history
2012Based in Mountain View, CA, US· patent location history
2014–2021Based in San Jose, CA, US· patent location history
2015–2021Patents assigned to MICROSOFT TECHNOLOGY LICENSING LLC (16)· patent assignee history
Inventor activity
Assignee and location history derived from the patent record.
Assignees
| Assignee | Patents | Span |
|---|---|---|
| MICROSOFT TECHNOLOGY LICENSING LLC | 16 | 2015–2021 |
| LEVIT MICHAELself | 6 | 2012–2015 |
| MICROSOFT CORP | 5 | 2012–2018 |
| BUNTSCHUH BRUCE MELVIN | 4 | 2012–2015 |
| CHANG SHUANGYU | 4 | 2012–2015 |
Locations
| Location | Patents | Span |
|---|---|---|
| San Jose, California, US | 18 | 2014–2021 |
| Berkeley, California, US | 2 | 2011–2012 |
| Mountain View, California, US | 1 | 2012 |