Search

Shubhabrata C Sengupta

from Menlo Park, CA
Age ~50
Get Report
Related to:
Harris Hen Hwang, 49Stephanie A Pressney, 63Caryn Klaff, 49Indrajit Sengupta, 51Alexandra S Klaff, 55Oren KlaffPushpa Sengupta, 51
Connected to:
Alpana Sengupta, 61Amalendu Sengupta, 79Ana L Sengupta, 79Anamika Sengupta, 52Anirban Sengupta, 53Anushree Sengupta, 44Arijit Sengupta, 51Asim Sengupta, 85Deborah J Sengupta, 52Dev Sengupta, 50

Shubhabrata Sengupta Phones & Addresses

  • 2125 Gordon Ave, Menlo Park, CA 94025 (530) 220-0464
  • Campbell, CA
  • Davis, CA
  • Houston, TX
  • Sunnyvale, CA

Resumes

Resumes

Shubhabrata Sengupta Photo 1

Shubhabrata Sengupta

View page

Publications

Us Patents

System, Method, And Computer Program Product For Converting A Reduction Algorithm To A Segmented Reduction Algorithm

View page
US Patent:
8321492, Nov 27, 2012
Filed:
Dec 11, 2008
Appl. No.:
12/333244
Inventors:
Shubhabrata Sengupta - Davis CA, US
Michael J. Garland - Lake Elmo MN, US
Assignee:
NVIDIA Corporation - Santa Clara CA
International Classification:
G06F 15/80
US Classification:
708620, 345505, 345501
Abstract:
A system, method, and computer program product are provided for converting a reduction algorithm to a segmented reduction algorithm. In operation, a reduction algorithm is identified. Additionally, the reduction algorithm is converted to a segmented reduction algorithm. Furthermore, the segmented reduction algorithm is performed to produce an output.

System, Method, And Computer Program Product For Converting A Scan Algorithm To A Segmented Scan Algorithm In An Operator-Independent Manner

View page
US Patent:
8243083, Aug 14, 2012
Filed:
Dec 11, 2008
Appl. No.:
12/333255
Inventors:
Michael J. Garland - Lake Elmo MN, US
Shubhabrata Sengupta - Davis CA, US
Assignee:
NVIDIA Corporation - Santa Clara CA
International Classification:
G06F 15/80
US Classification:
345505, 345419, 345501, 345506, 712 2, 712 10
Abstract:
A system, method, and computer program product are provided for converting a scan algorithm to a segmented scan algorithm in an operator independent manner. In operation, a scan algorithm and a limit index data structure are identified. Utilizing the limit index data structure, the scan algorithm is converted to a segmented scan algorithm in an operator-independent manner. Additionally, the segmented scan algorithm is performed to produce an output.

Deep Learning Models For Speech Recognition

View page
US Patent:
20190371298, Dec 5, 2019
Filed:
Aug 15, 2019
Appl. No.:
16/542243
Inventors:
- Sunnyvale CA, US
Carl CASE - San Francisco CA, US
Jared Casper - Sunnyvale CA, US
Bryan Catanzaro - Cupertino CA, US
Gregory Diamos - San Jose CA, US
Erich Elsen - Mountain View CA, US
Ryan Prenger - Oakland CA, US
Sanjeev Satheesh - Sunnyvale CA, US
Shubhabrata Sengupta - Menlo Park CA, US
Adam Coates - Sunnyvale CA, US
Andrew Ng - Mountain View CA, US
Assignee:
BAIDU USA LLC - Sunnyvale CA
International Classification:
G10L 15/06
Abstract:
Presented herein are embodiments of state-of-the-art speech recognition systems developed using end-to-end deep learning. In embodiments, the model architecture is significantly simpler than traditional speech systems, which rely on laboriously engineered processing pipelines; these traditional systems also tend to perform poorly when used in noisy environments. In contrast, embodiments of the system do not need hand-designed components to model background noise, reverberation, or speaker variation, but instead directly learn a function that is robust to such effects. A phoneme dictionary, nor even the concept of a “phoneme,” is needed. Embodiments include a well-optimized recurrent neural network (RNN) training system that can use multiple GPUs, as well as a set of novel data synthesis techniques that allows for a large amount of varied data for training to be efficiently obtained. Embodiments of the system can also handle challenging noisy environments better than widely used, state-of-the-art commercial speech systems.

Systems And Methods For A Multi-Core Optimized Recurrent Neural Network

View page
US Patent:
20170169326, Jun 15, 2017
Filed:
Apr 5, 2016
Appl. No.:
15/091413
Inventors:
- Sunnyvale CA, US
Awni Hannun - Palo Alto CA, US
Bryan Catanzaro - Cupertino CA, US
Dario Amodei - San Francisco CA, US
Erich Elsen - Mountain View CA, US
Jesse Engel - Oakland CA, US
Shubhabrata Sengupta - Menlo Park CA, US
Assignee:
Baidu USA LLC - Sunnyvale CA
International Classification:
G06N 3/04
G06N 3/08
Abstract:
Systems and methods for a multi-core optimized Recurrent Neural Network (RNN) architecture are disclosed. The various architectures affect communication and synchronization operations according to the Multi-Bulk-Synchronous-Parallel (MBSP) model for a given processor. The resulting family of network architectures, referred to as MBSP-RNNs, perform similarly to a conventional RNNs having the same number of parameters, but are substantially more efficient when mapped onto a modern general purpose processor. Due to the large gain in computational efficiency, for a fixed computational budget, MBSP-RNNs outperform RNNs at applications such as end-to-end speech recognition.

End-To-End Speech Recognition

View page
US Patent:
20170148431, May 25, 2017
Filed:
Nov 21, 2016
Appl. No.:
15/358102
Inventors:
- Sunnyvale CA, US
Jingdong Chen - Beijing, CN
Mike Chrzanowski - Sunnyvale CA, US
Erich Elsen - Mountain View CA, US
Jesse Engel - Oakland CA, US
Christopher Fougner - Palo Alto CA, US
Xu Han - Sunnyvale CA, US
Awni Hannun - Palo Alto CA, US
Ryan Prenger - Oakland CA, US
Sanjeev Satheesh - Sunnyvale CA, US
Shubhabrata Sengupta - Menlo Park CA, US
Dani Yogatama - Sunnyvale CA, US
Chong Wang - Redmond WA, US
Jun Zhan - San Jose CA, US
Zhenyao Zhu - Mountain View CA, US
Dario Amodei - San Francisco CA, US
Assignee:
Baidu USA LLC - Sunnyvale CA
International Classification:
G10L 15/06
G10L 25/18
G10L 15/197
G10L 15/16
Abstract:
Embodiments of end-to-end deep learning systems and methods are disclosed to recognize speech of vastly different languages, such as English or Mandarin Chinese. In embodiments, the entire pipelines of hand-engineered components are replaced with neural networks, and the end-to-end learning allows handling a diverse variety of speech including noisy environments, accents, and different languages. Using a trained embodiment and an embodiment of a batch dispatch technique with GPUs in a data center, an end-to-end deep learning system can be inexpensively deployed in an online setting, delivering low latency when serving users at scale.

Deployed End-To-End Speech Recognition

View page
US Patent:
20170148433, May 25, 2017
Filed:
Nov 21, 2016
Appl. No.:
15/358083
Inventors:
- Sunnyvale CA, US
Jingdong Chen - Beijing, CN
Mike Chrzanowski - Sunnyvale CA, US
Erich Elsen - Mountain View CA, US
Jesse Engel - Oakland CA, US
Christopher Fougner - Palo Alto CA, US
Xu Han - Sunnyvale CA, US
Awni Hannun - Palo Alto CA, US
Ryan Prenger - Oakland CA, US
Sanjeev Satheesh - Sunnyvale CA, US
Shubhabrata Sengupta - Menlo Park CA, US
Dani Yogatama - Sunnyvale CA, US
Chong Wang - Redmond WA, US
Jun Zhan - San Jose CA, US
Zhenyao Zhu - Mountain View CA, US
Dario Amodei - San Francisco CA, US
Assignee:
Baidu USA LLC - Sunnyvale CA
International Classification:
G10L 15/16
G10L 15/02
G10L 25/21
G10L 15/197
G10L 15/06
Abstract:
Embodiments of end-to-end deep learning systems and methods are disclosed to recognize speech of vastly different languages, such as English or Mandarin Chinese. In embodiments, the entire pipelines of hand-engineered components are replaced with neural networks, and the end-to-end learning allows handling a diverse variety of speech including noisy environments, accents, and different languages. Using a trained embodiment and an embodiment of a batch dispatch technique with GPUs in a data center, an end-to-end deep learning system can be inexpensively deployed in an online setting, delivering low latency when serving users at scale.

Systems And Methods For Speech Transcription

View page
US Patent:
20160171974, Jun 16, 2016
Filed:
Jun 9, 2015
Appl. No.:
14/735002
Inventors:
- Sunnyvale CA, US
Carl Case - San Francisco CA, US
Jared Casper - Sunnyvale CA, US
Bryan Catanzaro - Cupertino CA, US
Gregory Diamos - San Jose CA, US
Erich Elsen - Mountain View CA, US
Ryan Prenger - Oakland CA, US
Sanjeev Satheesh - Sunnyvale CA, US
Shubhabrata Sengupta - Menlo Park CA, US
Adam Coates - Sunnyvale CA, US
Andrew Y. Ng - Mountain View CA, US
Assignee:
BAIDU USA LLC - Sunnyvale CA
International Classification:
G10L 15/06
G10L 15/16
G10L 15/26
Abstract:
Presented herein are embodiments of state-of-the-art speech recognition systems developed using end-to-end deep learning. In embodiments, the model architecture is significantly simpler than traditional speech systems, which rely on laboriously engineered processing pipelines; these traditional systems also tend to perform poorly when used in noisy environments. In contrast, embodiments of the system do not need hand-designed components to model background noise, reverberation, or speaker variation, but instead directly learn a function that is robust to such effects. A phoneme dictionary, nor even the concept of a “phoneme,” is needed. Embodiments include a well-optimized recurrent neural network (RNN) training system that can use multiple GPUs, as well as a set of novel data synthesis techniques that allows for a large amount of varied data for training to be efficiently obtained. Embodiments of the system can also handle challenging noisy environments better than widely used, state-of-the-art commercial speech systems.
Control profile
Shubhabrata C Sengupta from Menlo Park, CA, age ~50 Get Report