Spoken Passphrase Step-Up Authentication via Separate Channel

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conversational interfaces lack robust authentication methods, with voice-print technologies being immature and spoken PINs or alphanumeric passwords being easily compromised, leading to inconsistent security in authorizing actions.

Innovation Solution

Implement step-up authentication using a spoken passphrase generated from a dictionary of easy-to-pronounce and distinguishable words, delivered via a separate communication channel (e.g., SMS, email, push notification) for asynchronous validation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a spoken PIN or alphanumeric password is used for authentication, then authentication capability is provided, but security is compromised as it can be easily overheared and compromised

Engineering Contradiction:
Improveauthentication securityVSAvoidoverhearing and compromise
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

A separate communication channel (SMS, email, push notification) acts as an intermediary to deliver the passphrase to the user's mobile device, separating the authentication flow from the conversational interface and preventing direct exposure of the password through the voice interface

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The authentication process is segmented into distinct phases: passphrase generation by the conversational interface, secure delivery through a separate communication channel to the user's mobile device, and verification when the user speaks the passphrase. This segmentation ensures that the password is never exposed through the conversational interface itself

Inventive Principle:
Principle #1Segmentation

2Reliability

If voice-print technologies are used for authentication, then authentication capability is provided, but the technology is not mature enough to be usable

Engineering Contradiction:
Improveauthentication capabilityVSAvoidtechnology maturity
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The system uses simple, easily deliverable one-time passphrases composed of common words rather than attempting to implement complex voice-print recognition. The authentication relies on the simplicity and ease of delivering short textual passphrases through standard communication channels rather than complex biometric verification

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

3Reliability

If a one-time password is used for authentication, then some level of security is provided, but the password must be alphanumeric which does not provide consistent results for voice interfaces

Engineering Contradiction:
Improvesecurity levelVSAvoidvoice interface consistency
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The passphrase is transformed from alphanumeric characters to a dictionary of pronounceable words, changing the parameter representation to be compatible with voice interfaces while maintaining security. The words are selected to be easy to pronounce and distinguishable, ensuring consistent voice recognition

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS12519773B2Step-up authentication using distinct communication channel delivery of spoken passphrase
Publication Date: 2026.01.06 TRUIST BANK
  • US12519773B2 patent drawing
  • US12519773B2 patent drawing
  • US12519773B2 patent drawing

AI summary

A system for step-up authentication using a spoken passphrase includes a conversational interface to: receive a conversational request for an action which requires step-up authorization; respond to the user through conversational artificial intelligence (AI) warning the user that an additional step is required; and a user mobile device to: receive, over a communication channel between a datacenter and the mobile device, to the user and different than the conversational interface, a secret passphrase generated using a spoken passphrase dictionary; wherein the conversational interface is further to receive a spoken submission from the user in response to the warning; wherein the datacenter confirms validity of the received spoken submission by comparing the spoken submission to the passphrase; and wherein the conversational interface is further to, in response to confirmation of the spoken submission, authorize the requested action.