AI Audio Signal Processing with Split Low-Latency Model Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio playback devices face limitations in executing complex audio signal processing due to low power and computation capabilities, leading to high latency when processing is performed in electronic devices and then transmitted to these devices.
Innovation Solution
A method involving a first electronic device obtaining an audio signal, transmitting it to a second electronic device for processing using multiple AI models with varying latencies, where the second device adjusts and processes the signal with lower latency before transmitting back to the first device for real-time reproduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If audio signal processing is performed in the audio playback device, then processing capability is improved, but power consumption increases and computation capability is insufficient
Solution Approach 1:
The audio signal processing task is divided into two segments: complex processing is performed in the electronic device (smartphone) while simple processing is performed in the audio playback device. This segmentation allows the complex AI-based processing to be done where computational power is available, while the playback device only handles lightweight tasks, thus resolving the contradiction between processing capability and power consumption.
Solution Approach 2:
The patent introduces an intermediary approach where the electronic device serves as a mediator for complex audio signal processing. The electronic device receives audio signals, performs complex processing using AI models, and transmits the processed signals to the audio playback device. This intermediary role allows the playback device to leverage the computational power of the electronic device without requiring high power consumption in itself.
2Productivity
If audio signal processing is performed in the electronic device and transmitted to the audio playback device, then processing capability is improved, but latency increases
Solution Approach 1:
The patent implements dynamic processing by adjusting the processing location based on the complexity of the audio signal and the available computational resources. The system dynamically switches between processing modes: complex processing in the electronic device when needed, and simple processing in the audio playback device when latency is critical. This dynamic approach resolves the contradiction between processing capability and latency by adapting the processing location in real-time.
Solution Approach 2:
The patent applies partial processing by performing only the necessary complex processing in the electronic device and leaving simple processing to be done in the audio playback device. This partial action approach reduces the overall processing time by avoiding unnecessary complex processing in the playback device, thus reducing latency while maintaining processing capability.
3Measurement precision
If complex audio signal processing techniques are used, then processing accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the processing complexity between two devices: the electronic device handles complex AI-based processing for high accuracy, while the audio playback device handles simple processing. This segmentation allows the system to achieve high processing accuracy through complex techniques in the electronic device without increasing the complexity of the audio playback device itself.
Solution Approach 2:
The patent uses a simplified copy of the processing function in the audio playback device. Instead of implementing the full complex AI processing in the playback device, it uses a simplified version that handles basic tasks. This copying approach allows the system to maintain high accuracy through the complex processing in the electronic device while keeping the playback device simple.
Data Source
AI summary
A method of processing an audio signal includes: obtaining an audio signal; transmitting the obtained audio signal to an external electronic device; receiving, from the external electronic device, second information for adjusting a third artificial intelligence model configured to process an audio signal in real time; inputting the received second information and the obtained audio signal to the third artificial intelligence model to adjust the third artificial intelligence model; inputting the obtained audio signal to the adjusted third artificial intelligence model to obtain a processed audio signal; and reproducing the processed audio signal.