Georgia Tech Research Corporation

Pindrop Security, Inc.

Vijay Balasubramaniyan

Michael Thomas Hunter

Aamir Poonawalla

Mustaque Ahamad

David Bryan Dewey

Kailash Patil

Scott Strong

Patrick Gerard Traynor

Raj Bandyopadhyay

Telvis Calhoun

Elie Khoury

Payas Gupta

Terry Nelms, Ii

Patrick Gerald Traynor

David A Dewey

Disclosed are systems and methods including computing-processes executing machine-learning architectures extract vectors representing disparate types of data and output predicted identities of users accessing computing services, without express identity assertions, and across multiple computing services, analyzing data from multiple modalities, for various user devices, and agnostic to architectures hosting the disparate computing service. The system invokes the identification operations of the machine-learning architecture, which extracts biometric embeddings from biometric data and context embeddings representing all or most of the types of metadata features analyzed by the system. The context embeddings help identify a subset of potentially matching identities of possible users, which limits the number of biometric-prints the system compares against an inbound biometric embedding for authentication. The types of extracted features originate from multiple modalities, including metadata from data communications, audio signals, and images. In this way, the embodiments apply a multi-modality machine-learning architecture.

Limiting identity space for voice biometric authentication

Various embodiments of the invention are detection systems and methods for detecting call provenance based on call audio. An exemplary embodiment of the detection system can comprise a characterization unit, a labeling unit, and an identification unit. The characterization unit can extract various characteristics of networks through which a call traversed, based on call audio. The labeling unit can be trained on prior call data and can identify one or more codecs used to encode the call, based on the call audio. The identification unit can utilize the characteristics of traversed networks and the identified codecs, and based on this information, the identification unit can provide a provenance fingerprint for the call. Based on the call provenance fingerprint, the detection system can identify, verify, or provide forensic information about a call audio source.

Systems and methods for detecting call provenance from call audio

Systems and methods for call detail record (CDR) analysis to determine a risk score for a call and identify fraudulent activity and for fraud detection in Interactive Voice Response (IVR) systems. An example method may store information extracted from received calls. Queries of the stored information may be performed to select data using keys, wherein each key relates to one of the received calls, and wherein the queries are parallelized. The selected data may be transformed into feature vectors, wherein each feature vector relates to one of the received calls and includes a velocity feature and at least one of a behavior feature or a reputation feature. A risk score for the call may be generated during the call based on the feature vectors.

Fraud detection in interactive voice response systems

Call detail record analysis to identify fraudulent activity

Growing community of inventors

Atlanta, GA, United States of America

Vijay Balasubramaniyan