Conference Audio Prioritization for Overlapping Speech Clarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video conferencing technologies face challenges with overlapping speech, leading to unintelligible communication and decreased clarity, especially as the number of participants increases, due to technical issues like diminished network bandwidth and latency.
Innovation Solution
A system and method that utilize a conference server to identify overlapping speech and selectively mute or prioritize audio based on factors such as speaking order, user information, topic relevance, and speaking frequency, ensuring that only one user's audio is provided in real-time while the other's audio is stored and provided later, optimizing communication clarity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If all participants' audio is transmitted simultaneously in a video conference, then all users can speak at the same time, but overlapping speech causes unintelligible communication and decreased clarity
Solution Approach 1:
The patent extracts and identifies overlapping speech segments from multiple audio streams using speech activity detection and overlap detection mechanisms. When overlap is detected, the system selectively mutes or prioritizes specific audio streams, extracting only the non-overlapping portions for transmission. This resolves the contradiction by maintaining concurrent speech capability while removing the harmful overlapping portions that degrade communication clarity.
Solution Approach 2:
The system dynamically adjusts audio stream transmission based on real-time detection of speech overlap. The audio processing is not static but adapts continuously by detecting when speech segments overlap and dynamically muting or prioritizing streams accordingly. This dynamic approach allows the system to maintain high communication clarity while preserving the ability for multiple users to speak concurrently when appropriate.
2Loss of information
If network bandwidth is increased to support multiple simultaneous audio streams, then all participants can communicate without overlap, but network resource consumption and latency increase
Solution Approach 1:
Instead of transmitting all audio streams simultaneously (excessive action), the system transmits only the necessary portions of audio streams that do not overlap (partial action). By using speech activity detection and overlap detection, the system transmits a subset of audio data that is sufficient for clear communication while avoiding the waste of network resources on redundant or conflicting audio segments. This resolves the contradiction by achieving adequate audio transmission quality without the excessive consumption of network bandwidth.
3Loss of information
If speech overlap detection and management is implemented, then communication clarity is improved, but system complexity increases
Solution Approach 1:
The audio processing system performs self-service by automatically detecting speech activity, identifying overlaps, and managing audio stream prioritization without requiring manual intervention or complex external control systems. The speech activity detection and overlap detection mechanisms enable the system to autonomously resolve speech conflicts, improving speech intelligibility while keeping the system complexity manageable through automated rather than manual processes.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems, methods, and non-transitory computer-readable media can be configured to determine first audio associated with a first user and second audio associated with a second user, the first user and the second user associated with a communication session. The second audio can be muted based on a determination that the first audio and the second audio overlap. The second audio can be provided based on completion of the first audio.