MSc thesis project proposal
[2023] Neuromorphic Real-Time Audio Denoiser
Audio denoising is a critical component in modern voice-controlled devices, digital assistants, and other audio-centric applications. The increasing demand for high-quality audio and low-power consumption necessitates the development of efficient and effective denoising techniques. This master's project aims to design, implement, and evaluate a neuromorphic hardware accelerator optimized for audio denoising tasks, taking advantage of the parallelism, low power consumption, and real-time processing capabilities of neuromorphic computing.
The project will involve a thorough investigation of existing audio denoising techniques, neuromorphic hardware architectures, and spiking neural networks (SNNs). The primary goal is to develop a hardware-software co-design approach that seamlessly integrates the neuromorphic accelerator with existing audio processing systems, ensuring compatibility and ease of deployment.
Assignment
- Conduct a literature review on audio denoising techniques, neuromorphic hardware architectures, and spiking neural networks.
- Analyze the computational and memory requirements of audio denoising algorithms to identify potential optimizations for neuromorphic hardware implementation.
- Design a novel neuromorphic hardware accelerator architecture optimized for audio denoising tasks.
- Implement a suitable spiking neural network (SNN) model for audio denoising, leveraging the capabilities of the proposed hardware accelerator.
- Develop a hardware-software co-design approach for integrating the neuromorphic accelerator with existing audio processing systems.
- Evaluate the performance of the proposed accelerator in terms of denoising quality, latency, energy consumption, and scalability, comparing it to traditional DSP and deep learning-based approaches.
- Document the project's findings and contribute to the development of a final report or publication.
Requirements
- Background in digital signal processing or audio processing.
- Familiarity with neuromorphic computing and hardware design.
- Knowledge of machine learning techniques, particularly spiking neural networks (SNNs).
- Proficiency in programming languages (e.g., Python, C++) and hardware description languages (e.g., VHDL, Verilog).
Contact
dr. Chang Gao
Electronic Circuits and Architectures Group
Department of Microelectronics
Last modified: 2023-03-22
