Artificial Neural Network Voice Signal Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio compression technologies in client computing devices are inefficient in reducing bandwidth and storage requirements for voice signals, as they do not effectively utilize advanced neural network architectures to minimize file size while maintaining quality.
Innovation Solution
An artificial neural network (ANN) is trained to compress voice signals by utilizing a compression portion from the input layer to a middle layer and decompressing using a decompression portion from the middle layer to the output layer, resulting in a lower file size and reduced bandwidth usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional audio compression technologies are used, then the voice signal can be transmitted and stored, but the file size remains large requiring more bandwidth and storage space
Solution Approach 1:
The patent replaces traditional mechanical audio compression algorithms with an artificial neural network system. The neural network learns optimal compression patterns through training, substituting conventional signal processing methods with a data-driven approach that achieves superior compression ratios while maintaining voice quality
Solution Approach 2:
The neural network is trained in advance on large datasets of voice signals to learn effective compression patterns before actual use. This preliminary training phase enables the network to perform rapid compression during operation without requiring real-time complex calculations, thus improving transmission speed
2Quantity of substance
If traditional audio compression technologies are used, then the voice signal can be transmitted, but bandwidth usage is not minimized
Solution Approach 1:
The neural network dynamically adjusts compression parameters based on the characteristics of the input voice signal. By learning from training data, the network optimizes compression ratios for different voice patterns, achieving minimal bandwidth usage while preserving essential voice quality parameters
Solution Approach 2:
The neural network creates an efficient compressed representation (copy) of the original voice signal that captures the essential characteristics. This compressed copy uses significantly less bandwidth while maintaining perceptual voice quality, as the network learns to preserve the most important signal features
Data Source
AI summary
In one embodiment, a method includes accessing a voice signal from a first user; compressing the voice signal using a compression portion of an artificial neural network trained to compress the first user's voice; and sending the compressed voice signal to a second client computing device.