GM GLOBAL TECHNOLOGY OPERATIONS LLC (20240296846). VOICE-BIOMETRICS BASED MITIGATION OF UNINTENDED VIRTUAL ASSISTANT SELF-INVOCATION simplified abstract

From WikiPatents
Jump to navigation Jump to search

VOICE-BIOMETRICS BASED MITIGATION OF UNINTENDED VIRTUAL ASSISTANT SELF-INVOCATION

Organization Name

GM GLOBAL TECHNOLOGY OPERATIONS LLC

Inventor(s)

Luke Heide of Plymouth MI (US)

Gaurav Talwar of Novi MI (US)

VOICE-BIOMETRICS BASED MITIGATION OF UNINTENDED VIRTUAL ASSISTANT SELF-INVOCATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240296846 titled 'VOICE-BIOMETRICS BASED MITIGATION OF UNINTENDED VIRTUAL ASSISTANT SELF-INVOCATION

Simplified Explanation

The patent proposes a solution using voice biometrics to prevent virtual assistant applications from self-invoking speech sessions.

  • Voice biometrics are used to create voiceprints based on synthetic speech utterances of the virtual assistant.
  • These voiceprints are stored in memory and compared with incoming audio streams to prevent self-invocation.
  • The processing system controls the virtual assistant and implements this voice biometrics-based solution.

Key Features and Innovation

  • Voice biometrics-based solution to prevent unwanted self-invocations by virtual assistant applications.
  • Creation of voiceprints using synthetic speech utterances for comparison with incoming audio streams.
  • Processing system controls the virtual assistant and implements the prevention mechanism.

Potential Applications

  • Virtual assistant applications in smart devices.
  • Call center automation systems.
  • Voice-controlled devices in various industries.

Problems Solved

  • Prevents virtual assistants from self-invoking speech sessions.
  • Enhances security and privacy by verifying the user's voiceprint.

Benefits

  • Improved user experience with virtual assistants.
  • Enhanced security measures for voice-controlled devices.
  • Prevention of accidental activations of virtual assistants.

Commercial Applications

Voice biometrics-based security solutions for virtual assistant applications in smart devices and call center automation systems.

Questions about Voice Biometrics

How does voice biometrics work in preventing self-invocations by virtual assistants?

Voice biometrics create voiceprints based on synthetic speech utterances to match with incoming audio streams, preventing self-invocation.

What are the potential drawbacks of using voice biometrics for security in virtual assistant applications?

Potential drawbacks may include false positives or difficulty in accurately matching voiceprints with incoming audio streams.


Original Abstract Submitted

a voice-biometrics based solution for unwanted self-invocations by virtual assistant applications is proposed. in various embodiments, a processing system is configured to control a virtual assistant. the processing system may have stored in a memory at least one voiceprint created using voice biometrics based on recorded utterances of synthetic speech from the virtual assistant. the at least one voiceprint may be used to prevent self-invocation of a virtual speech session by matching the voiceprint created using the synthetic speech utterances with the incoming audio stream.