公开(公告)号：US20240087593A1

公开(公告)日：2024-03-14

申请号：US17942289

申请日：2022-09-12

申请人： Nuance Communications, Inc.

发明人： Dushyant Sharma ,  Uwe Helmut JOST ,  Patrick Aubrey NAYLOR ,  Ljubomir MILANOVIC ,  William Francis GANONG, III

IPC分类号： G10L25/51 ,  G10L25/87 ,  H04S7/00

CPC分类号： G10L25/51 ,  G10L25/87 ,  H04S7/302 ,  H04S2420/01

摘要： A method, computer program product, and computing system for determining a plurality of transfer functions for a plurality of corresponding segments from a reference recording and a suspect recording. A delta transfer function between the plurality of transfer functions of a pair of corresponding segments of the plurality of corresponding segments is determined. A recording comparison confidence score is generated for the pair of corresponding segments based upon, at least in part, the delta transfer function. The suspect recording is verified based upon, at least in part, the plurality of recording comparison confidence scores.

公开(公告)号：US20240005908A1

公开(公告)日：2024-01-04

申请号：US18058266

申请日：2022-11-22

申请人： Nuance Communications, Inc.

发明人： Dushyant SHARMA ,  Patrick Aubrey NAYLOR ,  Ge LI

IPC分类号： G10L15/02 ,  G10L25/18

CPC分类号： G10L15/02 ,  G10L25/18

摘要： An acoustic environment profile estimation is provided for automatic speech recognition (ASR) to compensate for the acoustic behavior of an environment in which audio is collected. Examples receive an audio signal and extract spectral features and modulation features. Extracting spectral features involves determining Mel filter bank (MFB) coefficients, and extracting modulation features involves applying Fourier transforms. The spectral features and modulation features are combined, and an acoustic environment profile estimate is extracted and provided as an input to the ASR. In some examples, the acoustic environment profile estimate is realized as acoustic environment parameters, whereas in some other examples, the acoustic environment profile estimate is realized as an acoustic embedding vector. For versions using acoustic environment parameters, when the acoustic environment changes significantly, such as flooring changes and/or speakers or microphones changing position, a new set of acoustic environment parameters is determined.

公开(公告)号：US20240214404A1

公开(公告)日：2024-06-27

申请号：US18146620

申请日：2022-12-27

申请人： Nuance Communications, Inc.

发明人： Dushyant SHARMA ,  Patrick Aubrey NAYLOR ,  William Francis GANONG, III ,  Uwe Helmut JOST ,  Ljubomir MILANOVIC

IPC分类号： H04L9/40

CPC分类号： H04L63/1425 ,  H04L63/1491

摘要： A method, computer program product, and computing system for executing a plurality of requests to process data using a trained machine learning model. An anomalous pattern of requests including at least a threshold amount of out-of-domain data is identified from the plurality of requests. A potential model inversion attack is detected based upon, at least in part, identifying the anomalous pattern of requests.

公开(公告)号：US20240086759A1

公开(公告)日：2024-03-14

申请号：US17942276

申请日：2022-09-12

申请人： Nuance Communications, Inc.

发明人： Dushyant Sharma ,  Ljubomir MILANOVIC ,  Patrick Aubrey NAYLOR ,  Uwe Helmut JOST ,  William Francis GANONG, III

IPC分类号： G06N20/00 ,  G10L19/018

CPC分类号： G06N20/00 ,  G10L19/018

摘要： A method, computer program product, and computing system for identifying a target output token associated with an output of a machine learning model. A portion of training data corresponding to the target output token is modified with a watermark feature, thus defining watermarked training data.