Voice Signal Interruption Evaluation Using Power Sum
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional technology fails to accurately evaluate conversational impression related to interruption during simultaneous speech, as it relies on comparing overlapping time with a threshold, which is insufficient to detect the impression of interruption effectively.
Innovation Solution
An information processing apparatus and method that calculates the power of voice signals from multiple speakers, determines the continuation time when the combined power exceeds a threshold, and evaluates the conversational impression based on this continuation time, allowing for accurate assessment of interruption impressions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If overlapping time is compared with a threshold to evaluate conversation, then the evaluation process is simple, but the accuracy of detecting interruption impression is insufficient
Solution Approach 1:
The patent changes the evaluation parameter from simple overlapping time to a composite parameter that includes both overlapping time and voice power magnitude. This allows the system to detect not just whether speakers overlap, but also the relative volume levels, thereby accurately identifying interruption impressions where one speaker's voice significantly exceeds the other's during overlapping periods
2Device complexity
If only overlapping time is used to evaluate conversation, then the evaluation method is simple, but the conversational impression related to interruption cannot be evaluated
Solution Approach 1:
The patent merges two types of information: overlapping time duration and voice power magnitude ratio. By combining these parameters, the system preserves complete conversational impression information, including both the temporal aspect of interruption (how long speakers overlap) and the acoustic aspect (volume difference between speakers during overlap)
3Measurement precision
If voice power magnitude is considered in addition to overlapping time, then the accuracy of interruption detection is improved, but the calculation complexity increases
Solution Approach 1:
The patent introduces a power magnitude ratio parameter calculated as the ratio between the larger and smaller voice powers during overlapping speech. This single ratio parameter efficiently captures the volume difference information needed to identify interruption impressions, avoiding the need for complex multi-parameter analysis while maintaining high detection accuracy
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing apparatus includes a specifying unit configured to specify a first signal level of a first voice signal, a specifying unit configured to specify a second signal level of a second voice signal, and an evaluating unit configured to execute evaluation of at least one of the first voice signal and the second voice signal based on at least one of a sum of the first signal level and the second signal level and an average of the first signal level and the second signal level.