Directional Hearing Instrument Processing for Sound Source Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing directional signal processing in hearing instruments struggles to accurately distinguish between relevant and irrelevant sound sources in complex conversational situations, particularly in noisy environments like restaurants, leading to imprecise sound differentiation.
Innovation Solution
A method for directional signal processing in hearing instruments that utilizes two input transducers to detect the angular and orientation directions of sound sources relative to the wearer, employing filters and artificial neural networks to differentiate between parasitic and useful sound sources based on angular and orientation direction analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional directional signal processing is used in hearing instruments, then the device can process multiple input signals directionally, but the distinction between relevant and irrelevant sound sources becomes too rough or imprecise in complex conversational situations
Solution Approach 1:
The patent extends traditional 2D directional processing (azimuth angle only) to 3D spatial processing by incorporating elevation angle detection. This dimensional expansion enables more precise sound source localization and differentiation by utilizing vertical spatial information that was previously unavailable, thereby improving measurement precision without proportionally increasing processing complexity
Solution Approach 2:
The patent segments the sound processing task into distinct components: azimuth angle detection, elevation angle detection, and relevance classification. By dividing the complex signal processing into modular segments that can be independently optimized and processed, the system achieves higher precision in sound source differentiation while managing computational complexity through structured organization
2Reliability
If simple angular direction detection is used, then the processing is computationally efficient, but the ability to distinguish relevant sound sources in noisy environments like restaurants is insufficient
Solution Approach 1:
The patent replaces traditional mechanical signal processing methods with artificial neural network-based processing. The neural networks learn optimal patterns for identifying relevant sound sources from training data, enabling reliable distinction between useful and parasitic sounds in complex acoustic environments like restaurants, while the models can be efficiently deployed on hearing instrument hardware
Solution Approach 2:
The patent implements preliminary action by pre-training neural networks with large datasets of acoustic environments and sound source characteristics before deployment. This pre-processing and pre-learning phase enables the system to quickly and reliably identify relevant sound sources during actual use without requiring complex real-time computations, thus maintaining processing efficiency while improving reliability
Data Source
AI summary
A method performs directional signal processing for a hearing instrument. A first input signal is generated from an ambient sound by a first input transducer of the hearing instrument, and a second input signal is generated from the ambient sound by a second input transducer of the hearing instrument. On the basis of the first input signal and the second input signal, an angular direction of a sound source relative to a first reference direction, in particular to a frontal direction of a wearer of the hearing instrument, is detected at least approximately. On the basis of the first input signal and the second input signal, an orientation direction of the sound source, in particular relative to a second reference direction, is detected at least approximately.