Examining the Best Speech-to-Text Method for Audio Files in Podcasting

Siya Naik¹ , Avina Almeida² , Shubham Lotliker³

Section:Research Paper, Product Type: Journal-Paper
Vol.9 , Issue.5 , pp.25-29, Oct-2021

Online published on Oct 31, 2021

Copyright © Siya Naik, Avina Almeida, Shubham Lotliker . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
 

View this paper at   Google Scholar | DPI Digital Library

XML View     PDF Download

How to Cite this Paper

Abstract :
Podcasting is a great way to give insights or opinions on any topic to the audience. Podcasting requires both parties to be present physically at the location. But due to the pandemic crisis, this has caused a big problem. So, it is now carried out on an online platform. But the cons are the presence of noise in the audio files as well as miscommunication. The Spectral Gating method is used to remove the noise. This paper compares the various algorithms for converting audio to text by using various speech-to-text pretrained models. We performed an experiment on various audio files and the best accuracy rate was obtained for SpeechRecognition pretrained model.

Key-Words / Index Term :
Podcast; Subtitles; Spectral Gating; Speech-to-Text; Silero; Vosk; Mozilla DeepSpeech; SpeechRecognition; Word Error Rate; Accuracy.

References :
[1] Akhil Kanade, Sourabh Gune, Shubham Dharamkar, Rohan Gokhale, “Automatic Subtile Generation for Videos,” Interntional Journal of Enginneering Research and General Science, Vol.3, Issue.6, p.744,2015.
[2] Siya Sadashiv Naik, Gouri Bhatikar and Ugam Gaude, “Analysis of Best Algorithm for Noise Reduction in Podcasting,” Internatioonal Journal of Scientific Research in Science and Technology, Vol.8, Issue.3, pp24-249,2021.
[3] N Usha Rani, P N Girija, “Error Analysis to Improve the Speech Recogntion Accuracy on Telegu Language,” Indian Academic of Sciences, Vol.37.Part.6, p.747,2012.
[4] Aashish Agarwal, Torsten Zesch, “German End-to-end Speech Recognition based on DeepSpeech,” ResearchGate, Germany, Germany, pp.2-3, 2019.
[5] N. SelvaKumar, M. Rohini, C. Narmada, M. Yogeshprabhu, “Network Traffic Control Using AI,” International Journal of Scientific Research in Network Security and Communication, Vol.8, Issue.2, pp.13-21,2020.
[6] Muhammad Hafida Firmansyah, Anand Paul, Deblina Bhattachrya, Gul Malik Urfa, “A.I. based Emedded Speech to Text using DeepSpeech,” ResearchGate, South Korea, pp.1-5,2020.
[7] Dhara Bhatt, Bhargavi Khrishna, “Computer Assisted Pronounciation Learning System Using Speech Recognition Systems “PROnunciation Application”,” International Journal of Scientific Research in Computer Science and Engineering, Vol.7, Issue.6,pp.36-39,2019.