Call Intrusion Notification Timing via Speech Boundary Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing telecommunications features, such as whisper page and call waiting, often disrupt ongoing calls by injecting audible tones or tones immediately, causing misunderstandings, delays, and annoyance to participants, as they occur regardless of speech boundaries.
Innovation Solution
A computational device that uses a prosodic analyzer and silence analyzer to identify speech boundaries in voice streams, allowing notifications such as tones or messages to be played only at these boundaries, minimizing disruption by synchronizing them with pauses in speech or activity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If notifications are provided immediately upon detection, then response time is improved, but user annoyance and call disruption increase
Solution Approach 1:
The system performs preliminary analysis of the voice stream to identify speech boundaries and pauses before delivering notifications. The prosodic analyzer and silence analyzer prepare boundary information in advance, allowing notifications to be scheduled at optimal moments without delaying the notification itself, thus resolving the contradiction between immediate response and minimal disruption.
Solution Approach 2:
The notification delivery timing is made dynamic rather than fixed. The system continuously monitors speech boundaries and adjusts notification timing based on real-time speech patterns, delivering notifications during natural pauses or at speech boundaries. This dynamic adaptation allows the system to maintain quick response times while avoiding call disruption by adapting to the flowing speech pattern.
2Object-affected harmful factors
If notifications are provided at speech boundaries, then call disruption is reduced, but notification delivery time increases
Solution Approach 1:
The prosodic analyzer and silence analyzer operate continuously during the call, constantly analyzing the voice stream for speech boundaries and pauses. This continuous analysis ensures that when a notification needs to be delivered, the system can immediately identify the next appropriate speech boundary without delay, maintaining quick notification delivery while minimizing call disruption.
Solution Approach 2:
The system uses feedback from the continuous speech analysis to dynamically adjust notification timing. The analyzers provide real-time information about speech boundaries and pauses, which feeds back into the notification scheduling decision. This feedback mechanism allows the system to deliver notifications at optimal moments without significant delay, resolving the contradiction between reduced disruption and timely delivery.
3Adaptability or versatility
If traditional intrusion features are used, then communication flexibility is improved, but user experience deteriorates due to mid-sentence interruptions
Solution Approach 1:
The system applies different qualities to different parts of the call by analyzing local speech characteristics. Instead of uniformly interrupting calls, the system identifies specific locations (speech boundaries and pauses) where interruptions are least disruptive. This local quality approach maintains call intrusion capability while significantly improving user experience by avoiding mid-sentence interruptions.
Solution Approach 2:
The system changes the timing parameter of notification delivery based on speech analysis. Rather than using fixed timing, the delivery time is adjusted according to speech boundaries and pause duration. This parameter change allows the system to maintain communication flexibility while improving user experience by adapting notification timing to the natural rhythm of conversation.
Data Source
AI summary
A computational system is provided that includes:(a) a boundary analyzer 164, 168, and 216 operable to identify a next boundary in a predetermined activity; and(b) an interrupt handler 160 operable to (i) determine that a notification is to be provided to a user, (ii) determine that the user is currently engaged in the predetermined activity on a computational device, and (iii) provide the notification to the user in temporal proximity to the identified next boundary.