Voice Quality Enhancement via Statistical Audio Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice quality enhancement systems require a period of adjustment over multiple calls to stabilize and incur high costs due to the need for individual device characterization and setting value measurement, especially for new or upgraded devices.
Innovation Solution
A system that collects and classifies audio characteristic signals from electronic devices based on device characteristics, generating environment setting information to control audio characteristics for voice quality enhancement, reducing the need for separate adjustments and lowering measurement and construction costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a server side operator measures audio characteristic and prepares setting value one by one for various types of devices, then voice quality enhancement can be provided for each device, but construction cost and maintenance cost increase significantly
Solution Approach 1:
The patent uses statistical characteristics derived from analyzing audio data of multiple devices to create a unified setting value database. Instead of measuring each device individually, the system copies and generalizes patterns from sampled devices to apply to all devices of the same type, significantly reducing measurement costs while maintaining voice quality enhancement effectiveness.
Solution Approach 2:
The system changes the approach from individual device parameter measurement to statistical parameter analysis. By collecting audio characteristics from multiple devices and analyzing their statistical properties, the system derives universal setting values that can be applied across device populations, transforming the measurement process from time-consuming individual testing to efficient statistical sampling.
2Measurement precision
If a server side measures audio characteristic one by one for each device, then accurate setting values can be obtained, but the preparation period for new or upgraded devices increases
Solution Approach 1:
The system performs preliminary statistical analysis and setting value preparation in advance, before devices are deployed. By pre-collecting audio characteristics from representative devices and pre-calculating statistical setting values, the system eliminates the need for time-consuming individual measurements when new devices are introduced, allowing immediate application of pre-prepared setting values.
Solution Approach 2:
Instead of measuring each new device individually, the system copies and applies pre-derived statistical setting values that were calculated from analysis of multiple devices. This copying approach maintains measurement precision through statistical representation while dramatically reducing the preparation time required for new or upgraded devices.
3Reliability
If individual device adjustment is performed through multiple calls, then voice quality stabilizes, but the adjustment period extends over multiple calls
Solution Approach 1:
The system performs the adjustment action in advance by pre-calculating optimal setting values based on statistical analysis of audio characteristics. Instead of waiting for multiple calls to observe and adjust, the system pre-determines the settings that will stabilize voice quality, eliminating the need for extended adjustment periods while maintaining reliability.
Solution Approach 2:
The system uses feedback from analyzing audio characteristics of multiple devices to refine and update setting values in the database. This feedback mechanism allows the system to learn from real device performance and continuously improve the statistical setting values, achieving stable voice quality without requiring extended individual adjustment periods.
Data Source
AI summary
Provided is a method of controlling an audio characteristic of an electronic device performed by a server, the method including receiving, from a first electronic device, first audio characteristic information that is measured during a call of the first electronic device; receiving, from the first electronic device, first device characteristic information of the first electronic device; first classifying the first audio characteristic information based on the first device characteristic information; generating environment setting information based on classified audio characteristic information including the first classifying; receiving, from a second electronic device among the plurality of electronic devices, second device characteristic information of the second electronic device; and transmitting second environment setting information corresponding to the second device characteristic information to the second electronic device, wherein the second environment setting information is applied to control an audio characteristic for a call of the second electronic device.