Transfer Learning-Based Deep Residual Learning for Speech Recognition in Clean and Noisy Environments

TL;DR

This paper presents a transfer learning approach using residual neural networks to improve speech recognition accuracy in both clean and noisy environments, outperforming CNN and LSTM models.

Contribution

The study introduces a novel ResNet-based transfer learning framework for robust speech recognition in diverse acoustic conditions.

Findings

Achieved 98.94% accuracy in clean environments

Achieved 91.21% accuracy in noisy environments

Outperformed CNN and LSTM models in recognition accuracy

Abstract

Addressing the detrimental impact of non-stationary environmental noise on automatic speech recognition (ASR) has been a persistent and significant research focus. Despite advancements, this challenge continues to be a major concern. Recently, data-driven supervised approaches, such as deep neural networks, have emerged as promising alternatives to traditional unsupervised methods. With extensive training, these approaches have the potential to overcome the challenges posed by diverse real-life acoustic environments. In this light, this paper introduces a novel neural framework that incorporates a robust frontend into ASR systems in both clean and noisy environments. Utilizing the Aurora-2 speech database, the authors evaluate the effectiveness of an acoustic feature set for Mel-frequency, employing the approach of transfer learning based on Residual neural network (ResNet). The experimental results demonstrate a significant improvement in recognition accuracy compared to convolutional neural networks (CNN) and long short-term memory (LSTM) networks. They achieved accuracies of 98.94% in clean and 91.21% in noisy mode.