[2024] A Real-Time Neural Audio Denoising Chip

Audio denoising is critical 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 investigate state-of-the-art neural network-based audio denoising techniques and their efficient hardware implementation. The expected delivery will be an application-specific integrated circuit (ASIC) accelerating the audio denoising algorithm to achieve less than 10 millisecond latency.

Assignment

1. Conduct a literature review on audio denoising techniques and neural network hardware accelerators.
2. Analyze the computational and memory requirements of audio denoising algorithms to identify potential optimizations for hardware implementation.
3. Design and tape-out (or using FPGA) a novel neural network accelerator chip optimized for audio denoising tasks.

Requirements

1. Familiarity with digital circuit design using Verilog.
2. Knowledge of deep neural networks.
3. Experience with Python and PyTorch.

Contact

Last modified: 2024-06-11