公开(公告)号：US20240355337A1

公开(公告)日：2024-10-24

申请号：US18388364

申请日：2023-11-09

Applicant: PINDROP SECURITY, INC.

Inventor： Umair Altaf ,  Sai Pradeep Peri ,  Lakshay Phatela ,  Payas Gupta ,  Yitao Sun ,  Svetlana Afanaseva ,  Kailash Patil ,  Elie Khoury ,  Bradley Magnetta ,  Vijay Balasubramaniyan ,  Tianxiang Chen

IPC: G10L17/24

CPC classification number: G10L17/24

Abstract: Disclosed are systems and methods including software processes executed by a server that detect audio-based synthetic speech (“deepfakes”) in a call conversation. The server applies an NLP engine to transcribe call audio and analyze the text for anomalous patterns to detect synthetic speech. Additionally or alternatively, the server executes a voice “liveness” detection system for detecting machine speech, such as synthetic speech or replayed speech. The system performs phrase repetition detection, background change detection, and passive voice liveness detection in call audio signals to detect liveness of a speech utterance. An automated model update module allows the liveness detection model to adapt to new types of presentation attacks, based on the human provided feedback.

公开(公告)号：US11488605B2

公开(公告)日：2022-11-01

申请号：US16907951

申请日：2020-06-22

Applicant: PINDROP SECURITY, INC.

Inventor： Elie Khoury ,  Parav Nagarsheth ,  Kailash Patil ,  Matthew Garland

IPC: G10L17/02 ,  G10L17/04 ,  G10L25/24 ,  G10L17/18 ,  G10L19/02 ,  G10L17/06 ,  G10L17/00 ,  G10L25/51 ,  G10L25/30

Abstract: An automated speaker verification (ASV) system incorporates a first deep neural network to extract deep acoustic features, such as deep CQCC features, from a received voice sample. The deep acoustic features are processed by a second deep neural network that classifies the deep acoustic features according to a determined likelihood of including a spoofing condition. A binary classifier then classifies the voice sample as being genuine or spoofed.

公开(公告)号：US10553218B2

公开(公告)日：2020-02-04

申请号：US15709232

申请日：2017-09-19

Applicant: PINDROP SECURITY, INC.

Inventor： Elie Khoury ,  Matthew Garland

IPC: G10L17/00 ,  G10L17/18 ,  G10L17/02 ,  G10L17/06

Abstract: In a speaker recognition apparatus, audio features are extracted from a received recognition speech signal, and first order Gaussian mixture model (GMM) statistics are generated therefrom based on a universal background model that includes a plurality of speaker models. The first order GMM statistics are normalized with regard to a duration of the received speech signal. The deep neural network reduces a dimensionality of the normalized first order GMM statistics, and outputs a voiceprint corresponding to the recognition speech signal.

公开(公告)号：US20240363099A1

公开(公告)日：2024-10-31

申请号：US18388466

申请日：2023-11-09

Applicant: PINDROP SECURITY, INC.

Inventor： Umair Altaf ,  Sai Pradeep Peri ,  Lakshay Phatela ,  Payas Gupta ,  Yitao Sun ,  Svetlana Afanaseva ,  Kailash Patil ,  Elie Khoury ,  Bradley Magnetta ,  Vijay Balasubramaniyan ,  Tianxiang Chen

IPC: G10L15/02 ,  G10L15/16

CPC classification number: G10L15/02 ,  G10L15/16

Abstract: Disclosed are systems and methods including software processes executed by a server that detect audio-based synthetic speech (“deepfakes”) in a call conversation. The server applies an NLP engine to transcribe call audio and analyze the text for anomalous patterns to detect synthetic speech. Additionally or alternatively, the server executes a voice “liveness” detection system for detecting machine speech, such as synthetic speech or replayed speech. The system performs phrase repetition detection, background change detection, and passive voice liveness detection in call audio signals to detect liveness of a speech utterance. An automated model update module allows the liveness detection model to adapt to new types of presentation attacks, based on the human provided feedback.

公开(公告)号：US20240355319A1

公开(公告)日：2024-10-24

申请号：US18388385

申请日：2023-11-09

Applicant: PINDROP SECURITY, INC.

Inventor： Umair Altaf ,  Sai Pradeep Peri ,  Lakshay Phatela ,  Payas Gupta ,  Yitao Sun ,  Svetlana Afanaseva ,  Kailash Patil ,  Elie Khoury ,  Bradley Magnetta ,  Vijay Balasubramaniyan ,  Tianxiang Chen

IPC: G10L15/02 ,  G10L15/16

CPC classification number: G10L15/02 ,  G10L15/16

Abstract: Disclosed are systems and methods including software processes executed by a server that detect audio-based synthetic speech (“deepfakes”) in a call conversation. The server applies an NLP engine to transcribe call audio and analyze the text for anomalous patterns to detect synthetic speech. Additionally or alternatively, the server executes a voice “liveness” detection system for detecting machine speech, such as synthetic speech or replayed speech. The system performs phrase repetition detection, background change detection, and passive voice liveness detection in call audio signals to detect liveness of a speech utterance. An automated model update module allows the liveness detection model to adapt to new types of presentation attacks, based on the human provided feedback.

公开(公告)号：US11727942B2

公开(公告)日：2023-08-15

申请号：US17746832

申请日：2022-05-17

Applicant: Pindrop Security, Inc.

Inventor： Elie Khoury ,  Matthew Garland

IPC: G10L17/26 ,  H04L9/40 ,  G06F21/32 ,  G10L25/30 ,  G10L17/18 ,  G10L17/04 ,  G10L15/26 ,  G06V40/10 ,  G06V40/16 ,  G06F18/24 ,  G06V10/764 ,  G06V10/82 ,  G06V40/50

CPC classification number: G10L17/26 ,  G06F18/24 ,  G06F21/32 ,  G06V10/764 ,  G06V10/82 ,  G06V40/10 ,  G06V40/16 ,  G10L15/26 ,  G10L17/04 ,  G10L17/18 ,  G10L25/30 ,  H04L63/0861 ,  G06V40/178 ,  G06V40/50

Abstract: Systems and methods may generate, by a computer, a voice model for an enrollee based upon a set of one or more features extracted from a first audio sample received at a first time; receive at a second time a second audio sample associated with a caller; generate a likelihood score for the second audio sample by applying the voice model associated with the enrollee on the set of features extracted from the second audio sample associated with the caller, the likelihood score indicating a likelihood that the caller is the enrollee; calibrate the likelihood score based upon a time interval from the first time to the second time and at least one of: an enrollee age at the first time and an enrollee gender; and authenticate the caller as the enrollee upon the computer determining that the likelihood score satisfies a predetermined threshold score.

公开(公告)号：US11468901B2

公开(公告)日：2022-10-11

申请号：US16536293

申请日：2019-08-08

Applicant: PINDROP SECURITY, INC.

Inventor： Elie Khoury ,  Matthew Garland

IPC: G10L17/08 ,  G10L15/16 ,  G10L17/02 ,  G10L17/04 ,  G10L17/22 ,  G06N3/04 ,  G06N3/08 ,  G10L17/18

Abstract: The present invention is directed to a deep neural network (DNN) having a triplet network architecture, which is suitable to perform speaker recognition. In particular, the DNN includes three feed-forward neural networks, which are trained according to a batch process utilizing a cohort set of negative training samples. After each batch of training samples is processed, the DNN may be trained according to a loss function, e.g., utilizing a cosine measure of similarity between respective samples, along with positive and negative margins, to provide a robust representation of voiceprints.

公开(公告)号：US10854205B2

公开(公告)日：2020-12-01

申请号：US16505452

申请日：2019-07-08

Applicant: PINDROP SECURITY, INC.

Inventor： Elie Khoury ,  Matthew Garland

IPC: G10L17/20 ,  G10L17/18 ,  G10L17/02 ,  G10L17/04 ,  G10L19/028

Abstract: A system for generating channel-compensated features of a speech signal includes a channel noise simulator that degrades the speech signal, a feed forward convolutional neural network (CNN) that generates channel-compensated features of the degraded speech signal, and a loss function that computes a difference between the channel-compensated features and handcrafted features for the same raw speech signal. Each loss result may be used to update connection weights of the CNN until a predetermined threshold loss is satisfied, and the CNN may be used as a front-end for a deep neural network (DNN) for speaker recognition/verification. The DNN may include convolutional layers, a bottleneck features layer, multiple fully-connected layers and an output layer. The bottleneck features may be used to update connection weights of the convolutional layers, and dropout may be applied to the convolutional layers.

公开(公告)号：US10679630B2

公开(公告)日：2020-06-09

申请号：US16442368

申请日：2019-06-14

Applicant: PINDROP SECURITY, INC.

Inventor： Elie Khoury ,  Matthew Garland

IPC: G10L17/00 ,  H04M1/27 ,  G10L17/24 ,  G10L15/19 ,  G10L17/08 ,  G06N7/00 ,  G10L15/07 ,  G10L15/26 ,  G10L17/04

Abstract: Utterances of at least two speakers in a speech signal may be distinguished and the associated speaker identified by use of diarization together with automatic speech recognition of identifying words and phrases commonly in the speech signal. The diarization process clusters turns of the conversation while recognized special form phrases and entity names identify the speakers. A trained probabilistic model deduces which entity name(s) correspond to the clusters.

公开(公告)号：US10381009B2

公开(公告)日：2019-08-13

申请号：US15818231

申请日：2017-11-20

Applicant: PINDROP SECURITY, INC.

Inventor： Elie Khoury ,  Matthew Garland

IPC: G06N3/04 ,  G06N3/08 ,  G10L15/16 ,  G10L17/02 ,  G10L17/04 ,  G10L17/08 ,  G10L17/18 ,  G10L17/22

Abstract: The present invention is directed to a deep neural network (DNN) having a triplet network architecture, which is suitable to perform speaker recognition. In particular, the DNN includes three feed-forward neural networks, which are trained according to a batch process utilizing a cohort set of negative training samples. After each batch of training samples is processed, the DNN may be trained according to a loss function, e.g., utilizing a cosine measure of similarity between respective samples, along with positive and negative margins, to provide a robust representation of voiceprints.