No Detail Found!

Speech-to-Text Hybrid English to Yoruba SMS Translator

Author(s)
Akinbowale Nathaniel Babatunde Kwara State University,Malete,Nigeria
Ronke Seyi Babatunde Kwara State University,Malete,Nigeria
Bukola Fatimah Balogun Kwara State University,Malete,Nigeria
Afeez Adeshina Oke Federal College of Education,Iwo,Nigeria
Emmanuel Umar Tai Solarin University of Education,Ijebu Ode,Nigeria
Abstract
The necessity for quick translation from English to regional languages, such as Yoruba language arises from the fact that it is imperative for the widespread dissemination of knowledge and information. A speech-to-text enabled translation on Android devices and text-to-speech synthesizer (TTS) capable to deliver real-time translation of Yoruba language is presented. It increases the naturalness of oral communications, eases the communication, and furthers the understanding and use of Yoruba language among people of different backgrounds. This speech-to-text enabled translation system uses NLP on one hand and a TTS interpretation technique on the other. It uses an Android Google translation API synthesizer and a recognizer for English and Yoruba translation. The current study attempted to develop an English to Yoruba SMS translator with built-in text-to-speech and speech-to-text features. The Android Studio Integrated Development Environment was used to create the system (IDE). Whereas, a statistical machine translation algorithm was adopted for the implementation of Google API translation for English speech to text, TTS conversions, and Yoruba translations. About 100 computational rules were formulated for the Yoruba voice translation process. For the translation processes, a questionnaire was also created and circulated to 350 people. Approximately, 96% of those who responded provided input. The responses were subjected to SPSS statistical analysis. The results suggested that the translation system has a 92% accuracy which is comparable to that of a human (expert).
Publication Details
Page(s) 15-36
Published Journal: Innovative Computing Review (ICR), Volume: 4, Issue: 1, Year: 2024

Request the full text from PASTIC

Keywords
Machine Translation, natural language processing, Google translator
0

Citations

0

Downloads

50

Views