IP Library Granted Patent US 12665752
Granted Patent B2
US 12665752 · App. 18/947,048 · Granted Jun 23, 2026

Adaptive quantum-based voice over internet protocol (VoIP) security

Inventors: Adam Kristian King (Fort Mill, SC); James Siekman (Charlotte, NC); Andrew S. Thomas (Hunt Valley, MD); Jada Williams (Hunt Valley, MD); Kevin Sweeney (Newark, DE); Ilya Goussev (Kennesaw, GA); Dawn Turner (Addison, TX); Robert James Hepp (Hunt Valley, MD); Charlene Gorter (Charlotte, NC); Crystine J. Peebles (Newark, DE); Tiara Rurey Fry (Denver, CO); Clara Fritts (Washington, DC)
Assignee: Bank of America Corporation
H04L9/0852G10L17/02G10L21/0208H04L9/0819H04L9/0866H04L9/3231H04L63/0861
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12665752
App. No.
18/947,048
Granted
Jun 23, 2026
Kind
B2
Abstract

Systems, methods, and apparatus are provided for adaptive, quantum-based VoIP security. A biometric voiceprint may be generated from a VoIP communication. In response to failure to authenticate the voiceprint, the sound waves associated with the voiceprint may be selectively neutralized without terminating the VoIP communication. In response to authentication of the voiceprint, a biometric key may be generated based on the voiceprint. A quantum encryption key may be generated and distributed via a quantum channel. A VoIP session key may be generated based on the biometric encryption key and the quantum encryption key and used to encrypt the VoIP communication. An ambient sound associated with the VoIP communication may be isolated. In response to failure to authenticate the ambient sound, the caller may be required to provide an additional form of authentication.

Claims (97)

1 . A method for adaptive VOIP security, the method comprising:

receiving a biometric input comprising a voiceprint associated with a VOIP communication;

in response to failure to authenticate the voiceprint:

neutralizing sound waves associated with the voiceprint without terminating the VOIP communication; and

storing a record of the voiceprint in a fraud repository;

in response to authentication of the voiceprint:

generating a biometric encryption key based on the voiceprint;

at a quantum processor, generating a quantum encryption key comprising entangled photonic qubits;

based on input of the biometric encryption key and the quantum encryption key, generating a VoIP session key;

encrypting the VOIP communication with the VOIP session key;

isolating an ambient sound associated with the VOIP communication; and

in response to failure to authenticate the ambient sound, requiring an additional form of verification.

2 . The method of claim 1 , wherein authentication of the voiceprint comprises:

converting an analog audio signal to a digital voiceprint;

extracting voiceprint features comprising pitch, modulation, tone, and rhythm from the digital voiceprint; and

matching the voiceprint features to stored voiceprint features in an authentication database.

3 . The method of claim 1 , wherein neutralizing the sound waves comprises:

converting an audio signal into a set of frequencies;

identifying a frequency associated with an unauthenticated voiceprint; and

generating an inverse sound wave with the same amplitude and an opposite phase to the frequency associated with the unauthenticated voiceprint without affecting other frequencies in the set of frequencies.

4 . The method of claim 3 , wherein the inverse sound waves are generated at a low latency wave generator in real time.

5 . The method of claim 1 , wherein generating the quantum encryption key further comprises:

activating quantum key distribution protocols at a central server associated with the VOIP system;

distributing the quantum encryption key via a quantum channel; and

tuning the quantum encryption key using an error correction algorithm to correct for quantum noise.

6 . The method of claim 1 , wherein generating the VOIP session key comprises using both the biometric encryption key and the quantum encryption key as inputs to a key derivation function.

7 . The method of claim 6 wherein the key derivation function is an XOR operator.

8 . The method of claim 1 , wherein isolating the ambient sound comprises:

converting the ambient sound into a set of frequencies;

tuning the set of frequencies using an adaptive filtering algorithm;

at one or more AI/ML models, identifying an expected ambient sound and an anomaly sound in the set of frequencies;

filtering a frequency associated with the expected ambient sound and enhancing a frequency associated with the anomaly sound; and

storing an audio recording of the anomaly sound in a fraud repository.

9 . The method of claim 1 , wherein the expected ambient sound comprises one of office sounds, street sounds, vehicle sounds, and nature sounds.

10 . One or more non-transitory computer-readable media storing computer-executable instructions which, when executed by a processor on a computer system, perform a method for adaptive VOIP security, the method comprising:

receiving a biometric input comprising a voiceprint associated with a VOIP communication;

in response to failure to authenticate the voiceprint:

neutralizing sound waves associated with the voiceprint without terminating the VOIP communication; and

storing a record of the voiceprint in a fraud repository;

in response to authentication of the voiceprint:

generating a biometric encryption key based on the voiceprint;

generating a quantum encryption key comprising entangled photonic qubits;

generating a VOIP session key based on input of the biometric encryption key and the quantum encryption key;

encrypting the VOIP communication with the VOIP session key;

isolating an ambient sound associated with the VOIP communication; and

in response to failure to authenticate the ambient sound, requiring an additional form of verification.

11 . The media of claim 10 , wherein authentication of the voiceprint comprises:

converting an analog audio signal to a digital voiceprint;

extracting voiceprint features comprising pitch, modulation, tone, and rhythm from the digital voiceprint; and

matching the voiceprint features to stored voiceprint features in an authentication database.

12 . The media of claim 10 , wherein neutralizing the sound waves comprises:

converting an audio signal into a set of frequencies;

identifying a frequency associated with an unauthenticated voiceprint; and

generating an inverse sound wave with the same amplitude and an opposite phase to the frequency associated with the unauthenticated voiceprint without affecting other frequencies in the set of frequencies.

13 . The media of claim 10 , wherein generating the quantum encryption key further comprises:

activating quantum key distribution protocols at a central server associated with the VOIP system;

distributing the quantum encryption key via a quantum channel; and

tuning the quantum encryption key using an error correction algorithm to correct for quantum noise.

14 . The media of claim 10 , wherein generating the VOIP session key comprises using both the biometric encryption key and the quantum encryption key as inputs to a key derivation function.

15 . The media of claim 10 , wherein isolating the ambient sound comprises:

converting the ambient sound into a set of frequencies;

tuning the set of frequencies using an adaptive filtering algorithm;

at one or more AI/ML models, identifying an expected ambient sound and an anomaly sound in the set of frequencies;

filtering a frequency associated with the expected ambient sound and enhancing a frequency associated with the anomaly sound; and

storing an audio recording of the anomaly sound in a fraud repository.

16 . A system for adaptive VOIP security, the system comprising:

a quantum processor configured to generate a quantum encryption key comprising entangled qubits photonic qubits;

a processor associated with a VoIP system configured to:

receive a biometric input comprising a voiceprint associated with a VOIP communication;

in response to failure to authenticate the voiceprint:

neutralize sound waves associated with the voiceprint without terminating the VOIP communication; and

store a record of the voiceprint in a fraud repository;

in response to authentication of the voiceprint:

generate a biometric encryption key based on the voiceprint;

generate a VOIP session key based on input of the biometric encryption key and the quantum encryption key; and

encrypt the VOIP communication with the VOIP session key;

isolate an ambient sound associated with the VOIP communication; and

in response to failure to authenticate the ambient sound, require an additional form of verification.

17 . The system of claim 16 , wherein authentication of the voiceprint comprises:

converting an analog audio signal to a digital voiceprint;

extracting voiceprint features comprising pitch, modulation, tone, and rhythm from the digital voiceprint; and

matching the voiceprint features to stored voiceprint features in an authentication database.

18 . The system of claim 16 , wherein neutralizing the sound waves comprises:

converting an audio signal into a set of frequencies;

identifying a frequency associated with an unauthenticated voiceprint; and

generating an inverse sound wave with the same amplitude and an opposite phase to the frequency associated with the unauthenticated voiceprint without affecting other frequencies in the set of frequencies.

19 . The system of claim 16 , wherein generating the quantum encryption key further comprises:

activating quantum key distribution protocols at a central server associated with the VOIP system;

distributing the quantum encryption key via a quantum channel; and

tuning the quantum encryption key using an error correction algorithm to correct for quantum noise.

20 . The system of claim 16 , wherein synchronizing the biometric encryption key with the quantum encryption key comprises using both the biometric encryption key and the quantum encryption key as inputs to a key derivation function.

21 . The system of claim 16 , wherein isolating the ambient sound comprises:

converting the ambient sound into a set of frequencies;

tuning the set of frequencies using an adaptive filtering algorithm;

at one or more AI/ML models, identifying an expected ambient sound and an anomaly sound in the set of frequencies;

filtering a frequency associated with the expected ambient sound and enhancing a frequency associated with the anomaly sound; and

storing an audio recording of the anomaly sound in a fraud repository.