Sonova AG (20240323619). Processing chip for processing audio signals using at least one deep neural network in a hearing device simplified abstract
Processing chip for processing audio signals using at least one deep neural network in a hearing device
Organization Name
Inventor(s)
Samuel Bucher of Rapperswil (CH)
Hans-Ueli Roeck of Hombrechtikon (CH)
Fethi Cherigui of Adliswil (CH)
Processing chip for processing audio signals using at least one deep neural network in a hearing device - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240323619 titled 'Processing chip for processing audio signals using at least one deep neural network in a hearing device
The abstract describes a processing chip designed for processing audio signals using deep neural networks in a hearing device. The chip includes specialized compute units for processing different types of neural network layers, a control unit for coordinating the computation of each layer, a shared memory unit for storing data, and a data bus system for accessing the shared memory unit.
- The processing chip is optimized for processing audio signals using deep neural networks in a hearing device.
- It includes a first compute unit for processing convolutional neural network layers and a second compute unit for processing recurring neural network layers.
- A control unit directs the compute units on when to compute each layer of the deep neural network.
- A shared memory unit stores data to be processed in the neural network layers.
- A data bus system provides access to the shared memory unit for each compute unit.
Potential Applications: - Hearing aids with advanced audio processing capabilities - Speech recognition systems for hearing-impaired individuals - Noise cancellation technology for improved audio quality in noisy environments
Problems Solved: - Enhancing audio processing performance in hearing devices - Improving speech recognition accuracy for users with hearing loss - Reducing background noise interference in audio signals
Benefits: - Enhanced audio signal processing for better hearing device performance - Improved speech recognition accuracy for users with hearing impairments - Enhanced user experience in noisy environments with better noise cancellation
Commercial Applications: Title: Advanced Audio Processing Chip for Hearing Devices This technology can be used in the development of next-generation hearing aids with superior audio processing capabilities, speech recognition systems for individuals with hearing loss, and noise cancellation technology for improved audio quality in various environments. The market implications include increased demand for advanced hearing devices and improved user satisfaction with speech recognition and noise cancellation features.
Questions about the technology: 1. How does this processing chip improve audio signal processing in hearing devices? 2. What are the specific advantages of using deep neural networks in audio processing for hearing aids?
Original Abstract Submitted
a processing chip for processing audio signals using at least one deep neural network (dnn) in a hearing device comprises a first compute unit having a hardware architecture adapted for processing one or more convolutional neural network layers of the at least one dnn, a second compute unit having hardware architecture adapted for processing one or more recurring neural network layers of the at least dnn, a control unit for directing the first and second compute units when to compute a respective layer of the at least one dnn, a shared memory unit of storing data to be processed in respective layers of the at least one dnn, and a data bus system for providing access to the shared memory unit for each of the first and the second compute unit.
