Passive and continuous multi-speaker voice biometrics

AU2021254787B2Pending Publication Date: 2026-09-03PINDROP SECURITY INC
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
AU2021254787
Authority / Receiving Office
AU · AU
Patent Type
Applications
Current Assignee / Owner
Priority Date
2020-04-15
Filing Date
2021-04-15
Publication Date
2026-09-03
Estimated Expiration
2041-04-15

AI Technical Summary

Technical Problem

Current voice biometric systems face challenges in efficiently enrolling new speakers and differentiating between multiple speakers, especially in passive and dynamic environments, leading to issues with authentication accuracy and system configuration in IoT devices and over-the-top services.

Method used

A passive and continuous voice biometrics system that uses machine-learning models to extract embeddings from audio signals, generate similarity scores, and create or update speaker profiles without explicit user prompts, enabling unsupervised enrollment and recognition of speakers, even in the presence of multiple speakers.

Benefits of technology

This approach allows for seamless and accurate speaker identification and authentication across various applications, including IoT devices and call centers, by continuously updating profiles and reducing false acceptances and rejections, while maintaining user anonymity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000001_0000
    Figure 00000001_0000
  • Figure 00000108_0000
    Figure 00000108_0000
  • Figure 00000109_0000
    Figure 00000109_0000
Patent Text Reader

Abstract

Embodiments described herein provide for a voice biometrics system execute machine-learning architectures capable of passive, active, continuous, or static operations, or a combination thereof. Systems passively and / or continuously, in some cases in addition to actively and / or statically, enrolling speakers as the speakers speak into or around an edge device (e.g., car, television, radio, phone). The system identifies users on the fly without requiring a new speaker to mirror prompted utterances for reconfiguring operations. The system manages speaker profiles as speakers provide utterances to the system. Machine-learning architectures implement a passive and continuous voice biometrics system, possibly without knowledge of speaker identities. The system creates identities in an unsupervised manner, sometimes passively enrolling and recognizing known or unknown speakers. The system offers personalization and security across a wide range of applications, including media content for over-the-top services and IoT devices (e.g., personal assistants, vehicles), and call centers.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Training speech recognition systems using word sequences

    US10388272B1

  • Speaker recognition in the call center

    US20180082689A1

  • Voice identification feature optimization and dynamic registration methods, client, and server

    US20180366125A1