A Yoruba Language Automatic Speech Recognition System using Deep Learning Approach

Ogunjobi Olumide Jospeh; Agbonifo Oluwatoyin Catherine; Akinwonmi Akintoba E.

Research Article

A Yoruba Language Automatic Speech Recognition System using Deep Learning Approach

by Ogunjobi Olumide Jospeh, Agbonifo Oluwatoyin Catherine, Akinwonmi Akintoba E.

International Journal of Computer Applications

Foundation of Computer Science (FCS), NY, USA

Volume 187 - Issue 124

Published: July 2026

Authors: Ogunjobi Olumide Jospeh, Agbonifo Oluwatoyin Catherine, Akinwonmi Akintoba E.

10.5120/ijca0dcb3a1b147c

PDF

Ogunjobi Olumide Jospeh, Agbonifo Oluwatoyin Catherine, Akinwonmi Akintoba E. . A Yoruba Language Automatic Speech Recognition System using Deep Learning Approach. International Journal of Computer Applications. 187, 124 (July 2026), 60-65. DOI=10.5120/ijca0dcb3a1b147c

                        @article{ 10.5120/ijca0dcb3a1b147c,
                        author  = { Ogunjobi Olumide Jospeh,Agbonifo Oluwatoyin Catherine,Akinwonmi Akintoba E. },
                        title   = { A Yoruba Language Automatic Speech Recognition System using Deep Learning Approach },
                        journal = { International Journal of Computer Applications },
                        year    = { 2026 },
                        volume  = { 187 },
                        number  = { 124 },
                        pages   = { 60-65 },
                        doi     = { 10.5120/ijca0dcb3a1b147c },
                        publisher = { Foundation of Computer Science (FCS), NY, USA }
                        }

                        %0 Journal Article
                        %D 2026
                        %A Ogunjobi Olumide Jospeh
                        %A Agbonifo Oluwatoyin Catherine
                        %A Akinwonmi Akintoba E.
                        %T A Yoruba Language Automatic Speech Recognition System using Deep Learning Approach%T 
                        %J International Journal of Computer Applications
                        %V 187
                        %N 124
                        %P 60-65
                        %R 10.5120/ijca0dcb3a1b147c
                        %I Foundation of Computer Science (FCS), NY, USA

Abstract

In this paper, a Yorùbá Language Automatic Speech Recognition System capable of recognizing words spoken by the users based on preprocessed stored data was designed and implemented. Dataset from Data Science and Computational Research Laboratory, Federal University of Technology, Akure, Nigeria, were made used of. Speech feature sequences were extracted using Mel-frequency cepstral coefﬁcients (MFCC) technique. Kaggle environment was deployed, where python programming language was used to implement the Transformer and LSTM (Long Short-Term Memory) models. The preprocessed data were fed into both models for training, both models were tested and evaluated across the following standard evaluation metrics WER, MER and CER. The results obtained from both models shows a promising approach that could be adopted for Yorùbá Language speech recognition system. The research work describes ASR for standard Yorúbà language. The source language and target language is Yorùbá language, with focus on speaking in Yorùbá language (voice data) and the output in Yorùbá language text form (text data).

References

IBM Cloud Education (2020). Natural Language Processing (NLP). Retrieved from https://www.ibm.com/cloud/learn/natural-language-processing
Brownlee, J. (2019). Deep learning for Natural Language Processing (NLP). Retrieved from https://machinelearningmastery.com/natural-langage-processing/
IBM Cloud Education (2020). What is Speech Recognition?. Retrieved from https://www.ibm.com/topics/speech-recognition
IGI Global (2021). What is ASR. Retrieved from https://www.igiglo.com/dictionary/asr/1566
Zajechowski, M. (2021). Automatic Speech Recognition (ASR) Software: An Introduction. Retrieved from. https://usabilitygeek.com/automatic-speech-recognition-asr-software-an-introduction/
Raut and Deoghare (2016). Automatic Speech Recognition and its Applications. IRJET, 03(05), 2368-2371
Adetunmbi, O. A., Obe, O. O., and Iyanda, J. N. (2016). Development of Standard Yorùbá speech-to-text system using HTK. International Journal of Speech Technology, 19(4), 92-94.
Abiola, O.B., Adeyemo O.A., Saka-Balogun O.Y., and Okesola F. (2020). A Web-Based Yorùbá to English Bilingual Lexicon for Building Technicians. International Journal of Advanced Trends in Computer Science and Engineering, 9(1), 793- 800
Eberhard, D. M., Gary F. S., and Fennig, C. D. (2021). Ethnologue: Languages of the World. Twenty-fourth. Retrieved from. https://www.ethnologue.com/language/yor
Okanlawon, J. (2016). An Analysis of the Yorùbá Language with English Phonetics, Phonology, Morphology and Syntax. Retrieved from. https://cos.northeastern.edu/wp-content/uploads/2018/09/Jolaade-Okanlawon-An-Analysis-of-Yorùbá-with-English.pdf
Ibiyemi, T.S., and Akintola, A.G. (2012). Automatic Speech Recognition for Telephone Voice Dialing in Yorùbá. International Journal of Engineering Research & Technology (IJERT), 1(4), 1-6
Zhang, Y., and Yzhang5, S. I. (2013). Speech Recognition Using Deep Learning Algorithms, pp. 1-5
Modupe, I. A, Sefara, T. J, and Ojo, S.O. (2019). Yorùbá Gender Recognition from Speech using Attention-based BiLSTM. In Proceedings of the First International Workshop on NLP Solutions for Under Resourced Languages (NSURL) co-located with ICNLSP. Short papers, pp. 16-22
Dong, L., Xu, S. and Xu, B. (2018). Speech-Transformer: A No-Recurrence Sequence-To-Sequence Model for Speech Recognition. International Conference on Acoustics, Speech and signal processing, pp. 5884-5888

Index Terms

Computer Science

Information Sciences

No index terms available.

Keywords

Automatic Speech Recognition (ASR) Mel-Frequency Cepstral Coefficients (MFCC) Speech-to-Text Low-Resource Language Deep Learning