Adaptive EVS Audio Codec Switching for RF Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The Enhanced Voice Services (EVS) Full Band (EVS-FB) codec provides superior sound quality but is less efficient and unreliable in certain scenarios, particularly when voice is the only sound involved or in poor radio frequency (RF) conditions, leading to resource wastage and potential voice quality issues.
Innovation Solution
A technique for adaptively switching between different EVS audio codecs during a communication session, using machine learning to determine the presence of non-vocal frequencies, such as music, to selectively employ the EVS-FB codec for enhanced audio bandwidth when beneficial and less-than-full-band codecs when not necessary, and transitioning to more efficient codecs in poor RF conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If EVS-FB codec is used to provide extended audio bandwidth up to 20 kHz, then sound quality for music and other sounds is significantly improved, but resource consumption increases and efficiency decreases
Solution Approach 1:
The system dynamically switches between EVS-FB and other EVS codecs based on real-time detection of non-vocal frequencies. When music or other extended frequency content is detected, EVS-FB is activated to provide high-quality audio. When only voice is present, the system transitions to more efficient codecs, thereby adapting resource consumption to actual audio content requirements.
Solution Approach 2:
The invention changes the audio bandwidth parameter from fixed to variable. By using machine learning models to analyze audio content and detect non-vocal frequencies, the system adjusts the audio bandwidth parameter dynamically - utilizing full 20 kHz bandwidth only when necessary for music content, and reducing to narrower bandwidths for voice-only content, thus optimizing the trade-off between sound quality and resource consumption.
2Manufacturing precision
If EVS-FB codec is used for extended audio bandwidth, then music and sound quality is improved, but reliability decreases in poor RF conditions
Solution Approach 1:
The system dynamically adapts codec selection based on both audio content and RF conditions. When poor RF conditions are detected, the system automatically switches from EVS-FB to more robust codecs that are better suited for voice-only transmission, thereby maintaining call reliability. This dynamic adaptation ensures that the system prioritizes connection stability over extended bandwidth when RF conditions deteriorate.
3Adaptability or versatility
If EVS-FB codec is used continuously, then extended audio bandwidth is available, but network efficiency decreases
Solution Approach 1:
The invention implements variable audio bandwidth parameter adjustment based on content analysis. By using machine learning models to detect the presence of non-vocal frequencies, the system changes the audio bandwidth parameter dynamically - utilizing EVS-FB's full 20 kHz capability only when music or other extended frequency content is detected, and switching to narrower bandwidth modes for voice-only content, thereby optimizing network efficiency while maintaining adaptability.
Solution Approach 2:
The system employs self-service mechanisms through automated machine learning-based audio content analysis and codec selection. The device autonomously determines when EVS-FB is needed by analyzing audio input for non-vocal frequencies, eliminating the need for manual intervention or continuous high-bandwidth allocation, thus improving network efficiency while maintaining audio quality when needed.
Data Source
AI summary
Described herein are techniques, devices, and systems for providing an optimal voice experience over varying radio frequency (RF) conditions while using EVS audio codecs. A user equipment (UE) may adaptively transition between using a music-capable EVS codec (e.g., EVS-FB) as a default audio codec that provides a first audio bandwidth and a different EVS audio codec that provides a second audio bandwidth that is less than the first audio bandwidth. The transition to the different EVS audio codec may occur in response to determining a value indicative of a RF condition associated with a serving base station is less than a threshold value, which allows for providing preserving at least a minimal level of voice quality in degraded RF conditions.