This paper is published in Volume-6, Issue-3, 2020
Area
Computer Science Engineering
Author
Gorli Harshini
Co-authors
Gayatri Yasaswini Pappala, Gorle Manasa, Gollamandala Sam Shalini, M. Sion Kumari
Org/Univ
Andhra University College of Engineering for Women, Visakhapatnam, Andhra Pradesh, India
Pub. Date
23 May, 2020
Paper ID
V6I3-1299
Publisher
Keywords
Deep convolutional neural networks, Interpolation method, Sub-pixel convolution, Bottleneck architecture.

Citationsacebook

IEEE
Gorli Harshini, Gayatri Yasaswini Pappala, Gorle Manasa, Gollamandala Sam Shalini, M. Sion Kumari. Audio super resolution using Neural Networks, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
Gorli Harshini, Gayatri Yasaswini Pappala, Gorle Manasa, Gollamandala Sam Shalini, M. Sion Kumari (2020). Audio super resolution using Neural Networks. International Journal of Advance Research, Ideas and Innovations in Technology, 6(3) www.IJARIIT.com.

MLA
Gorli Harshini, Gayatri Yasaswini Pappala, Gorle Manasa, Gollamandala Sam Shalini, M. Sion Kumari. "Audio super resolution using Neural Networks." International Journal of Advance Research, Ideas and Innovations in Technology 6.3 (2020). www.IJARIIT.com.

Abstract

In this era of technological advancement, people tend to demand high-quality videos, audios, and images. So deep convolutional neural networks play an important role in learning low-resolution data and obtaining high-resolution data by performing interpolation method. This is similar to the image super-resolution. Here we introduce a signal processing technique to convert the low resolution into high-resolution data with the help of signal processing methods such as up-sampling and down-sampling using subpixel convolution through Bottleneck architecture. Our model tests the missing values within the low-resolution signals and forms high-resolution signals. This technique is applied to telephony, upscaling, text to speech conversion, and also for investigations in many departments. We test the effect of convolutions used in the signal processing and measures the compatibility and scalability for the generative model of audio.
Paper PDF