公开(公告)号：US20220076665A1

公开(公告)日：2022-03-10

申请号：US17016360

申请日：2020-09-09

Applicant: International Business Machines Corporation

Inventor： Aaron K. Baughman ,  Corey B. Shelton ,  Stephen C. Hammer ,  Shikhar Kwatra

IPC: G10L15/16 ,  G10L21/0272 ,  G10L25/03 ,  G06F3/16 ,  G06N3/08

Abstract: The disclosure includes using dilation of speech content from a separated audio input for speech recognition. An audio input from a speaker and predicted changes for the audio input based on an external noise are received at a CNN (Convolutional Neural Network). In the CNN, diarization is applied to the audio input to predict how a dilation of speech content from the speaker changes the audio input to generate a CNN output. A resulting dilation is determined from the CNN output. A word error rate is determined for the dilated CNN output to determine an accuracy for speech to text outputs. An adjustment parameter is set to change a range of the dilation based on the word error rate, and the resulting dilation of the CNN output is adjusted based on the adjustment parameter to reduce the word error rate.

公开(公告)号：US11056104B2

公开(公告)日：2021-07-06

申请号：US15605984

申请日：2017-05-26

Applicant: International Business Machines Corporation

Inventor： Aaron K. Baughman ,  Stephen C. Hammer ,  Mauro Marzorati

IPC: G10L15/183 ,  G10L15/06 ,  G10L15/00

Abstract: In an approach for acoustic modeling with a language model, a computer isolates an audio stream. The computer identifies one or more language models based at least in part on the isolated audio stream. The computer selects a language model from the identified one or more language models. The computer creates a text based on the selected language model and the isolated audio stream. The computer creates an acoustic model based on the created text. The computer generates a confidence level associated with the created acoustic model. The computer selects a highest ranked language model based at least in part on the generated confidence level.

公开(公告)号：US10938942B2

公开(公告)日：2021-03-02

申请号：US16366057

申请日：2019-03-27

Applicant: International Business Machines Corporation

Inventor： Aaron K. Baughman ,  Brian M. O'Connell ,  David Alexander Provan ,  Stephen C. Hammer

IPC: H04L29/08 ,  G06N7/00 ,  G06N20/00

Abstract: Determining and/or adjusting cache expire times (sometimes herein referred to as “cache expires”) based, at least in part to how much excitement a live event, such as a live sporting event, is generating in its audience. After the excitement from the live event has dissipated, the cache expires can be reset to the values that they would otherwise have during normal operations.

公开(公告)号：US20210012809A1

公开(公告)日：2021-01-14

申请号：US17033933

申请日：2020-09-28

Applicant: International Business Machines Corporation

Inventor： Aaron K. Baughman ,  Stephen C. Hammer ,  Gray Cannon

IPC: G11B27/036 ,  G06N3/08 ,  G06K9/42 ,  G06K9/00 ,  H04N21/25 ,  H04N21/234 ,  H04N21/8549 ,  H04N21/845 ,  G06K9/62

Abstract: Techniques for padding audiovisual clips (for example, audiovisual clips of sporting events) for the purpose of causing the clip to have a predetermined duration so that the padded clip can be evaluated for viewer interest by a machine learning (ML) algorithm. The unpadded clip is padded with audiovisual segment(s) that will cause the padded clip to have a level of viewer interest that it would have if the unpadded clip had been longer. In some embodiments the padded segments are synthetic images generated by a generative adversarial network such that the synthetic images would have the same level of viewer interest (as adjudged by an ML algorithm) as if the unpadded clip had been shot to be longer.

公开(公告)号：US20200314199A1

公开(公告)日：2020-10-01

申请号：US16366057

申请日：2019-03-27

Applicant: International Business Machines Corporation

Inventor： Aaron K. Baughman ,  Brian M. O'Connell ,  David Alexander Provan ,  Stephen C. Hammer

IPC: H04L29/08 ,  G06N20/00 ,  G06N7/00

Abstract: Determining and/or adjusting cache expire times (sometimes herein referred to as “cache expires”) based, at least in part to how much excitement a live event, such as a live sporting event, is generating in its audience. After the excitement from the live event has dissipated, the cache expires can be reset to the values that they would otherwise have during normal operations.

公开(公告)号：US20200296177A1

公开(公告)日：2020-09-17

申请号：US16354674

申请日：2019-03-15

Applicant: International Business Machines Corporation

Inventor： Stephen C. Hammer ,  Gray Cannon ,  Aaron K. Baughman

IPC: H04L29/08 ,  G06F16/43 ,  G06Q30/02

Abstract: Techniques for tailoring sampling rates for data from interactive digital properties on a feature-by-feature basis and collecting the data using the tailored sampling rates. Each feature may have an independent sampling rate irrespective of sampling rates assigned to other features. The independent sampling rates are determined based on at least one factor of predictive feature usage information based on historical feature usage information, predetermined rules, and current usage velocity of the feature. In some embodiments the independent sampling rate is influenced by the usage of an allocated resource provided to the digital property relative to a total allocation of that resource for a given time period. In some embodiments, the allocated resource is server calls to a digital data analytics server for the purposes of providing feature usage information from the interactive digital property for the performance of digital data analytics.

公开(公告)号：US10595101B2

公开(公告)日：2020-03-17

申请号：US15921653

申请日：2018-03-15

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Michele Merler ,  Dhiraj Joshi ,  Quoc-Bao Nguyen ,  Stephen C. Hammer ,  John Joseph Kent ,  John R. Smith ,  Rogerio Feris

IPC: H04N21/44 ,  H04N21/442 ,  H04N21/466 ,  H04N21/431 ,  H04N21/439 ,  H04N21/8549 ,  G06N20/00 ,  G06N3/08 ,  G06N3/04 ,  G06K9/00

Abstract: A method and system for auto-curating a media are provided. Media content is received over the network interface. A set of markers is identified for the media content, each marker corresponding to one of a plurality of visible and audible cues in the media content. Segments in the media content are identified based on the identified set of markers. An excitement score is computed for each segment based on the identified markers that fall within the segment. A highlight clip is generated by identifying segments having excitement scores greater than a threshold.

公开(公告)号：US20190289372A1

公开(公告)日：2019-09-19

申请号：US15921653

申请日：2018-03-15

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Michele Merler ,  Dhiraj Joshi ,  Quoc-Bao Nguyen ,  Stephen C. Hammer ,  John Joseph Kent ,  John R. Smith ,  Rogerio Feris

IPC: H04N21/8549 ,  G06N3/08 ,  G06N3/04 ,  G06N99/00 ,  H04N21/439 ,  H04N21/44 ,  H04N21/431 ,  H04N21/466 ,  G06K9/00 ,  H04N21/442

Abstract: A method and system for auto-curating a media are provided. Media content is received over the network interface. A set of markers is identified for the media content, each marker corresponding to one of a plurality of visible and audible cues in the media content. Segments in the media content are identified based on the identified set of markers. An excitement score is computed for each segment based on the identified markers that fall within the segment. A highlight clip is generated by identifying segments having excitement scores greater than a threshold.

公开(公告)号：US20180342239A1

公开(公告)日：2018-11-29

申请号：US15605984

申请日：2017-05-26

Applicant: International Business Machines Corporation

Inventor： Aaron K. Baughman ,  Stephen C. Hammer ,  Mauro Marzorati

IPC: G10L15/183 ,  G10L15/06 ,  G10L15/00 ,  G10L25/51 ,  G10L15/18

Abstract: In an approach for acoustic modeling with a language model, a computer isolates an audio stream. The computer identifies one or more language models based at least in part on the isolated audio stream. The computer selects a language model from the identified one or more language models. The computer creates a text based on the selected language model and the isolated audio stream. The computer creates an acoustic model based on the created text. The computer generates a confidence level associated with the created acoustic model. The computer selects a highest ranked language model based at least in part on the generated confidence level.

公开(公告)号：US20180270286A1

公开(公告)日：2018-09-20

申请号：US15463506

申请日：2017-03-20

Applicant: International Business Machines Corporation

Inventor： Aaron K. Baughman ,  Stephen C. Hammer ,  Christopher E. Holladay ,  Mauro Marzorati

IPC: H04L29/06 ,  H04L29/08 ,  H04L12/26

CPC classification number: H04L67/22 ,  H04L65/4076 ,  H04L65/601 ,  H04L67/18 ,  H04W4/029 ,  H04W40/244

Abstract: Focus data of a remote user is analyzed to determine a focus shift from a first area to a second area at an event arena. A beacon density is computed at the second area, where the beacon density includes a number of physical beacons corresponding to a number of local users at the second area, a number of virtual beacons corresponding to a number of remote users focused on the second area, or a combination of thereof. When the beacon density at the second area exceeds a threshold density, an instruction to a streaming source is generated. The streaming source is caused to change a streaming content, to form changed streaming content that is related to the second area.