VoIP Audio Quality Assessment Using Receiver Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for estimating the Mean Opinion Score (MOS) in VoIP systems are unreliable and intrusive, as they rely on network status monitoring rather than direct analysis of audio data, failing to accurately reflect the quality of experience due to the disconnect between network conditions and human perception.
Innovation Solution
A non-intrusive audio quality assessment system that analyzes VoIP call statistics, including packet loss concealment and acceleration/slowdown module impacts, to estimate MOS by determining influence factors and using codec-bitrate reference values, providing a more accurate and real-time quality assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional network status monitoring is used to estimate MOS, then the estimation process is simple to implement, but the accuracy and reliability of MOS estimation deteriorates because network status is not directly related to human hearing perception
Solution Approach 1:
The patent replaces the conventional network status monitoring approach with an audio signal processing approach. Instead of using network parameters (packet loss rate, jitter) to estimate MOS, the system directly analyzes the received audio signal characteristics (energy, zero-crossing rate, spectral features) to determine MOS. This substitution of the estimation mechanism from network-layer to application-layer audio analysis resolves the contradiction by making the estimation directly related to human hearing perception while maintaining computational feasibility through standardized audio processing algorithms.
2Reliability
If conventional network status monitoring is used to estimate MOS, then the implementation is non-intrusive to the audio stream, but the reliability of MOS estimation worsens due to the disconnect between network conditions and human perception
Solution Approach 1:
The system enables the audio quality assessment to be self-performed by analyzing the received audio signal itself. The receiver device extracts features directly from the decoded audio stream and autonomously determines MOS without requiring external network status data or manual intervention. This self-service approach improves reliability by using the actual audio output that the user hears, while the automated feature extraction and MOS calculation maintain operational simplicity.
3Measurement precision
If audio data analysis is performed in each time window to provide accurate MOS estimation, then the accuracy of MOS estimation is improved, but the processing complexity and computational load increases
Solution Approach 1:
The patent divides the continuous audio stream into discrete time windows (e.g., 20ms frames) and performs MOS estimation independently for each window. This segmentation allows the system to use efficient short-time Fourier transform (STFT) and frame-based feature extraction, reducing computational complexity compared to analyzing the entire audio stream at once. The segmented approach enables real-time processing with lower power consumption while maintaining high accuracy through frequent, localized assessments.
Solution Approach 2:
The system extracts only the most critical audio features (energy, zero-crossing rate, spectral centroid, bandwidth) necessary for MOS estimation rather than performing complete audio analysis. By focusing on the essential features that most strongly correlate with perceived quality, the system achieves accurate MOS estimation with reduced computational effort and lower processing power requirements.
Data Source
AI summary
A new audio quality assessment system includes an assessment system running in a receiver system of a VoIP communication system. The new audio quality assessment system determines an accurate MOS of a VoIP call within a time window. The audio quality assessment system determines an effective PLC counter, a PLC impact factor, an effective AS counter, an AS impact factor, a network impact factor, a codec type of the received voice packets, a bitrate of the received voice packets, an initial MOS from a configured codec-bitrate MOS table, and determines the accurate MOS based on these data. The determined MOS is more accurate and efficiently obtained since it is based on efficiently collected statistics of the receiver system's modules and a pre-configured codec-bitrate MOS table.