Effects of Dataset Sampling Rate for Noise Cancellation through Deep Learning

TL;DR

This study investigates how different dataset sampling rates affect the performance of deep neural networks, specifically ConvTasNET, in noise cancellation tasks suitable for mobile devices, highlighting a trade-off between audio quality and processing time.

Contribution

It demonstrates that training ConvTasNET on higher sampling rate datasets improves noise cancellation quality but increases processing time, informing optimal sampling choices for mobile applications.

Findings

Higher sampling rates (48kHz) improve audio quality metrics.

Processing time increases with higher sampling rates.

ConvTasNET trained at 48kHz outperforms lower rates in noise suppression.

Abstract

Background: Active noise cancellation has been a subject of research for decades. Traditional techniques, like the Fast Fourier Transform, have limitations in certain scenarios. This research explores the use of deep neural networks (DNNs) as a superior alternative. Objective: The study aims to determine the effect sampling rate within training data has on lightweight, efficient DNNs that operate within the processing constraints of mobile devices. Methods: We chose the ConvTasNET network for its proven efficiency in speech separation and enhancement. ConvTasNET was trained on datasets such as WHAM!, LibriMix, and the MS-2023 DNS Challenge. The datasets were sampled at rates of 8kHz, 16kHz, and 48kHz to analyze the effect of sampling rate on noise cancellation efficiency and effectiveness. The model was tested on a core-i7 Intel processor from 2023, assessing the network's ability to produce clear audio while filtering out background noise. Results: Models trained at higher sampling rates (48kHz) provided much better evaluation metrics against Total Harmonic Distortion (THD) and Quality Prediction For Generative Neural Speech Codecs (WARP-Q) values, indicating improved audio quality. However, a trade-off was noted with the processing time being longer for higher sampling rates. Conclusions: The Conv-TasNET network, trained on datasets sampled at higher rates like 48kHz, offers a robust solution for mobile devices in achieving noise cancellation through speech separation and enhancement. Future work involves optimizing the model's efficiency further and testing on mobile devices.