A New Residual Convolutional Neural Network-Based Speech Improvement
Journal Title: International Journal of Innovative Research in Computer Science and Technology - Year 2022, Vol 10, Issue 6
Abstract
Among the most crucial methods for denoising a noisy voice signal and enhancing its quality is speech enhancement. This study makes use of Adaptive Residual Neural Network technique to reduces maximum off background noise. This method continuously monitors the background noise depends upon the environmental changes using SNR parameter. It has two functions first one is non linear functions followed by convolutional neural networks and second one is linearity followed due to Residual neural networks. By using these factors remove background noise even SNR is low conditions. Compared to other techniques this technique is new,fastest, requires less training and also reduces size.
Authors and Affiliations
M. Balasubrahmanyam, B. Haribabu, and P. Uday Kumar
Keywords
Related Articles
- EP ID EP746007
- DOI 10.55524/ijircst.2022.10.6.7
- Views 44
- Downloads 0
Deep Web Crawler: A Review
In today’s scenario, there is an ample amount of data on the internet that can be accessed by everyone. This is the data that can be indexed by search engines. There are softwares named Web Crawlers that explore the WWW...
Hybrid Active Power Filter for Power Quality Improvement
A Deadbeat current controller for an LC-coupling hybrid active power filter is proposed, which can track with the reference compensation current with low steady- state error and fast dynamic response. Moreover, it can le...
Methods for Performance Evaluation of the Small Enterprises in Bulgaria
The Evaluation of the work performance of the personnel is one of the key activities in human resources management. In this article the methods of evaluation of performance that are applied in the small enterprises in Bu...
AI Based Food Quality Recommendation System
Deep learning has been evidenced to be a cutting-edge technology for big data scrutiny with a huge figure of effective cases in image processing, speech recognition, object detection, and so on. Lately, it has also been...
Minimizing Keystrokes/Clicks In Traditional Desktop Interfaces: Introducing A Modality-Diversification Layer
Desktop software applications are built with traditional user interfaces in mind. Unlike modern Smartphones and tablets where Tactile and other haptic user-interface modalities are primary, desktop user-interfaces put ex...