Binaural Hearing Devices Sharing Neural Networks to Cut Battery Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hearing devices face limitations in computational resources and battery capacity, which hinder efficient and flexible audio signal processing, particularly in binaural scenarios.
Innovation Solution
A hearing device arrangement that transmits neural network data between connected hearing devices to distribute processing tasks, allowing for shared computational power, reduced battery consumption, and improved binaural processing by incorporating spatial information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If neural network processing is executed on each hearing device independently, then audio signal processing quality can be maintained, but computational resources and battery capacity are quickly depleted
Solution Approach 1:
The patent divides the neural network processing tasks between two hearing devices. Instead of each device independently executing the complete neural network for audio signal processing, the processing is segmented such that each device executes only a portion of the neural network operations. This reduces the computational burden and energy consumption on each individual device while maintaining overall processing quality through collaboration between the pair of devices.
2Productivity
If computational power is increased to improve audio signal processing, then processing quality improves, but device size and complexity increase
Solution Approach 1:
The patent merges the computational capabilities of two hearing devices to achieve high-level audio signal processing. By combining the processing power of both devices and enabling them to execute neural network operations collaboratively, the system achieves enhanced processing capability without requiring each individual device to have high computational resources. This merging approach allows complex neural network processing while keeping individual device complexity low.
3Productivity
If all neural network processing is performed on one hearing device, then processing efficiency improves, but the other device cannot benefit from spatial information
Solution Approach 1:
The patent uses the communication link between the two hearing devices as an intermediary to exchange neural network data and spatial information. Each device processes local audio signals and exchanges intermediate results through this intermediary channel. This allows both devices to contribute their spatial information and computational results, ensuring that no spatial information is lost while maintaining processing efficiency through distributed computation.
Data Source
AI summary
A hearing device arrangement includes two hearing devices which are connected to each other in a data transmitting manner. Each hearing device includes an audio input unit for obtaining an input audio signal, a processing unit for audio signal processing of the input audio signal to obtain an output audio signal, a neural network which, when executed by the processing unit performs a processing step of the audio signal processing, and an audio output unit for outputting the output audio signal. The hearing device arrangement is configured to transmit neural network data of the neural network of at least one of the hearing devices to the respective other hearing device to be used in the audio signal processing by the processing unit of the respective other hearing device.