Voicemail Recall via Speaker Verification and TTS Masking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current voicemail systems do not allow users to recall messages from remote voicemail systems, especially when the caller is not a member of the voicemail system or uses a phone without voicemail support, limiting the ability to retract unintended messages.
Innovation Solution
A method and system that uses speaker verification through text-to-speech (TTS) to authenticate callers by extracting and comparing voice signatures stored with the voicemail message, allowing non-members to recall messages from remote voicemail systems by playing extracted words and masking the voice signature, enabling deletion of the message upon authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If speaker verification with TTS masking is implemented to enable recall by non-members, then accessibility and versatility are improved, but device complexity and security risks increase
Solution Approach 1:
The patent introduces a TTS masking layer as an intermediary between the voice signature verification system and the caller. This mediator converts challenge words into speech that masks the original voice signature, allowing non-member callers to be authenticated through spoken challenges without exposing the underlying voice biometric data, thus enabling recall functionality while managing system complexity
Solution Approach 2:
The patent replaces traditional mechanical authentication methods (passwords, PINs) with voice-based speaker verification. The system captures voice prints during message delivery and uses acoustic signal processing to verify caller identity, substituting conventional authentication mechanics with biometric verification to enable recall for non-members
2Reliability
If voice signature extraction and storage are implemented for all messages, then security and reliability are improved, but loss of information and storage requirements increase
Solution Approach 1:
The patent extracts only the essential voice signature characteristics from complete voice messages during delivery. By separating and storing only the biometric voice print data rather than entire messages, the system achieves reliable authentication while minimizing information loss and protecting caller privacy through selective data extraction
Solution Approach 2:
The patent applies different quality levels to different parts of voice data. Complete voice messages are processed to extract high-quality voice signatures for authentication, while the original messages remain intact for playback. This local quality differentiation ensures authentication reliability without permanently storing sensitive voice biometric information in a way that compromises privacy
3Reliability
If TTS masking is used to protect voice signature during authentication, then security is improved, but measurement precision and authentication time increase
Solution Approach 1:
The patent performs preliminary voice signature extraction and storage during the initial message delivery phase. By pre-processing and storing voice biometric data before the recall authentication occurs, the system eliminates the need for real-time voice processing during authentication, maintaining both security through TTS masking and measurement precision through pre-established voice profiles
Solution Approach 2:
The patent uses TTS masking that provides more security coverage than strictly necessary by masking the entire voice signature rather than just critical portions. This excessive masking action enhances security against sophisticated attacks while the pre-extracted voice signatures maintain authentication precision by comparing only essential biometric features
4Ease of operation
If recall functionality is extended to non-members using any phone, then ease of operation is improved, but loss of information and system control increase
Solution Approach 1:
The patent implements feedback mechanisms where the system monitors and logs all recall attempts, including caller identification, timestamp, and authentication results. This feedback loop allows the system to maintain control by tracking message recall events while enabling easy operation for authenticated callers, balancing accessibility with information control through comprehensive auditing
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables users to recall voicemail messages from any phone, ensuring only the original sender can delete the message, enhancing message control and security while allowing recall functionality across different voicemail systems.
Implementation Method 1
playing the extracted words to the caller using text-to-speech, TTS, to mask the voice signature of the user
Implementation Method 2
extracting a voice signature of the user upon receiving the voicemail message from the user
Implementation Method 3
authenticating the caller using speaker verification by extracting words from the voicemail message in the target mailbox
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Techniques and systems for recalling voicemail messages from remote voicemail systems are disclosed. In one embodiment, a method for recalling a voicemail message from a target mailbox can include: accessing a voicemail system by a caller using a device; authenticating the caller using speaker verification; and deleting the voicemail message from the target mailbox. The target mailbox owner can be a member of the voicemail system, while the caller can be a non-member of that voicemail system. The device may be configured to support a telephony user interface (TUI), for example.