公开(公告)号：US11817117B2

公开(公告)日：2023-11-14

申请号：US17162907

申请日：2021-01-29

Applicant: NVIDIA Corporation

Inventor： Utkarsh Vaidya ,  Ravindra Yeshwant Lokhande ,  Viraj Gangadhar Karandikar ,  Niranjan Rajendra Wartikar ,  Sumit Kumar Bhattacharya

IPC: G10L25/78 ,  G10L25/30 ,  G06N3/02

CPC classification number: G10L25/78 ,  G06N3/02 ,  G10L25/30 ,  G10L2025/786

Abstract: In various examples, end of speech (EOS) for an audio signal is determined based at least in part on a rate of speech for a speaker. For a segment of the audio signal, EOS is indicated based at least in part on an EOS threshold determined based at least in part on the rate of speech for the speaker.

公开(公告)号：US20220246167A1

公开(公告)日：2022-08-04

申请号：US17162907

申请日：2021-01-29

Applicant: NVIDIA Corporation

Inventor： Utkarsh Vaidya ,  Ravindra Yeshwant Lokhande ,  Viraj Gangadhar Karandikar ,  Niranjan Rajendra Wartikar ,  Sumit Kumar Bhattacharya

IPC: G10L25/78 ,  G10L25/30 ,  G06N3/02

Abstract: In various examples, end of speech (EOS) for an audio signal is determined based at least in part on a rate of speech for a speaker. For a segment of the audio signal, EOS is indicated based at least in part on an EOS threshold determined based at least in part on the rate of speech for the speaker.

公开(公告)号：US20240311080A1

公开(公告)日：2024-09-19

申请号：US18676243

申请日：2024-05-28

Applicant: NVIDIA Corporation

Inventor： Utkarsh Vaidya ,  Sumit Bhattacharya

IPC: G06F3/16 ,  G06N3/045 ,  G06N7/01

CPC classification number: G06F3/165 ,  G06F3/162 ,  G06N3/045 ,  G06N7/01

Abstract: The disclosure is directed to a process that can predict and prevent an audio artifact from occurring. The process can monitor the systems, processes, and execution threads on a larger system/device, such as a mobile or in-vehicle device. Using a learning algorithm, such as deep neural network (DNN), the information collected can generate a prediction of whether an audio artifact is likely to occur. The process can use a second learning algorithm, which also can be a DNN, to generate recommended system adjustments that can attempt to prevent the audio glitch from occurring. The recommendations can be for various systems and components on the device, such as changing the processing system frequency, the memory frequency, and the audio buffer size. After the audio artifact has been prevented, the system adjustments can be reversed fully or in steps to return the system to its state prior to the system adjustments.

公开(公告)号：US20210358490A1

公开(公告)日：2021-11-18

申请号：US16876433

申请日：2020-05-18

Applicant: Nvidia Corporation

Inventor： Utkarsh Vaidya ,  Sumit Bhattacharya ,  Viraj Karandikar ,  Niranjan Wartikar

IPC: G10L15/197 ,  G10L15/04 ,  G10L15/02 ,  G10L15/16 ,  G10L15/22 ,  G06N3/08

Abstract: Apparatuses, systems, and techniques are presented to recognize speech in an audio signal. In particular, various embodiments can indicate an end of one or more speech segments based, at least in part, on one or more characters predicted to be within these one or more speech segments.

公开(公告)号：US11567728B2

公开(公告)日：2023-01-31

申请号：US17121373

申请日：2020-12-14

Applicant: Nvidia Corporation

Inventor： Utkarsh Vaidya ,  Sumit Bhattacharya

IPC: G06F3/16 ,  G06N7/00 ,  G06N3/04

Abstract: The disclosure is directed to a process that can predict and prevent an audio artifact from occurring. The process can monitor the systems, processes, and execution threads on a larger system/device, such as a mobile or in-vehicle device. Using a learning algorithm, such as deep neural network (DNN), the information collected can generate a prediction of whether an audio artifact is likely to occur. The process can use a second learning algorithm, which also can be a DNN, to generate recommended system adjustments that can attempt to prevent the audio glitch from occurring. The recommendations can be for various systems and components on the device, such as changing the processing system frequency, the memory frequency, and the audio buffer size. After the audio artifact has been prevented, the system adjustments can be reversed fully or in steps to return the system to its state prior to the system adjustments.

公开(公告)号：US10896021B2

公开(公告)日：2021-01-19

申请号：US16285941

申请日：2019-02-26

Applicant: Nvidia Corporation

Inventor： Utkarsh Vaidya ,  Sumit Bhattacharya

IPC: G06F3/16 ,  G06N7/00 ,  G06N3/04

Abstract: The disclosure is directed to a process that can predict an audio glitch, and then attempt to preempt the audio glitch. The process can monitor the systems, processes, and execution threads on a larger system or device, such as a mobile device or an in-vehicle device. Using a learning algorithm, such as deep neural network (DNN), the information collected can generate a prediction of whether an audio glitch is likely to occur. An audio glitch can be an audio underrun condition. The process can use a second learning algorithm, which also can be a DNN, to generate recommended system adjustments that can attempt to prevent the audio glitch from occurring. The recommendations can be for various systems and components on the device, such as changing the processing system frequency, the memory frequency, and the audio buffer size. After the audio underrun condition has abated, the system adjustments can be reversed fully or in steps to return the system to its state prior to the system adjustments.

公开(公告)号：US11995378B2

公开(公告)日：2024-05-28

申请号：US18161326

申请日：2023-01-30

Applicant: NVIDIA Corporation

Inventor： Utkarsh Vaidya ,  Sumit Bhattacharya

IPC: G06F3/16 ,  G06N3/045 ,  G06N7/01

CPC classification number: G06F3/165 ,  G06F3/162 ,  G06N3/045 ,  G06N7/01

Abstract: The disclosure is directed to a process that can predict and prevent an audio artifact from occurring. The process can monitor the systems, processes, and execution threads on a larger system/device, such as a mobile or in-vehicle device. Using a learning algorithm, such as deep neural network (DNN), the information collected can generate a prediction of whether an audio artifact is likely to occur. The process can use a second learning algorithm, which also can be a DNN, to generate recommended system adjustments that can attempt to prevent the audio glitch from occurring. The recommendations can be for various systems and components on the device, such as changing the processing system frequency, the memory frequency, and the audio buffer size. After the audio artifact has been prevented, the system adjustments can be reversed fully or in steps to return the system to its state prior to the system adjustments.

公开(公告)号：US20230298579A1

公开(公告)日：2023-09-21

申请号：US18202228

申请日：2023-05-25

Applicant: NVIDIA Corporation

Inventor： Utkarsh Vaidya ,  Sumit Bhattacharya ,  Viraj Karandikar ,  Niranjan Wartikar

IPC: G10L15/197 ,  G10L15/04 ,  G10L15/02 ,  G10L15/22 ,  G06N3/08 ,  G10L15/16

CPC classification number: G10L15/197 ,  G10L15/04 ,  G10L15/02 ,  G10L15/22 ,  G06N3/08 ,  G10L15/16 ,  G10L2015/223

Abstract: Apparatuses, systems, and techniques are presented to recognize speech in an audio signal. In particular, various embodiments can indicate an end of one or more speech segments based, at least in part, on one or more characters predicted to be within these one or more speech segments.

公开(公告)号：US20230168857A1

公开(公告)日：2023-06-01

申请号：US18161326

申请日：2023-01-30

Applicant: NVIDIA Corporation

Inventor： Utkarsh Vaidya ,  Sumit Bhattacharya

IPC: G06F3/16 ,  G06N3/045 ,  G06N7/01

CPC classification number: G06F3/165 ,  G06F3/162 ,  G06N3/045 ,  G06N7/01

Abstract: The disclosure is directed to a process that can predict and prevent an audio artifact from occurring. The process can monitor the systems, processes, and execution threads on a larger system/ device, such as a mobile or in-vehicle device. Using a learning algorithm, such as deep neural network (DNN), the information collected can generate a prediction of whether an audio artifact is likely to occur. The process can use a second learning algorithm, which also can be a DNN, to generate recommended system adjustments that can attempt to prevent the audio glitch from occurring. The recommendations can be for various systems and components on the device, such as changing the processing system frequency, the memory frequency, and the audio buffer size. After the audio artifact has been prevented, the system adjustments can be reversed fully or in steps to return the system to its state prior to the system adjustments.

公开(公告)号：US20210103425A1

公开(公告)日：2021-04-08

申请号：US17121373

申请日：2020-12-14

Applicant: Nvidia Corporation

Inventor： Utkarsh Vaidya ,  Sumit Bhattacharya

IPC: G06F3/16 ,  G06N7/00 ,  G06N3/04

Abstract: The disclosure is directed to a process that can predict and prevent an audio artifact from occurring. The process can monitor the systems, processes, and execution threads on a larger system/device, such as a mobile or in-vehicle device. Using a learning algorithm, such as deep neural network (DNN), the information collected can generate a prediction of whether an audio artifact is likely to occur. The process can use a second learning algorithm, which also can be a DNN, to generate recommended system adjustments that can attempt to prevent the audio glitch from occurring. The recommendations can be for various systems and components on the device, such as changing the processing system frequency, the memory frequency, and the audio buffer size. After the audio artifact has been prevented, the system adjustments can be reversed fully or in steps to return the system to its state prior to the system adjustments.