An Overview of Techniques for Speech-to-Text Conversion and Text Summarisation
AFRICAN JOURNAL OF APPLIED RESEARCH · 30.07.2026 · 12 (5) · 897-916
DOI: 10.26437/b58e1n19 ↗Kurz gefasst
Fachartikel in der Zeitschrift AFRICAN JOURNAL OF APPLIED RESEARCH, online erschienen am 30.07.2026. Verfasst von C. S. Gode und 7 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Purpose
The study seeks to critically review and synthesise existing approaches—ranging from traditional statistical and rule-based methods to contemporary deep learning and transformer-based architectures—used in converting spoken language into text Design / Methodology / Approach: This review paper evaluates conventional approaches and structural patterns registrable in recent innovations across both opuses, with a keen focus on their complexities, centrality, and scope of application.
Findings
Several strengths and weaknesses of these methods are assessed in terms of precision, time, and flexibility across different languages and trends within the same languages. Two types of text summarisation techniques are compared, namely extractive and abstractive text summarisation, emphasising advances by transformer models such as BERT and GPT in generating context-sensitive summaries. Research Limitation: The evaluation primarily focuses on conventional approaches and recent innovations in speech-to-text conversion and text summarisation, with an emphasis on their computational complexity, model-centric architectures, and application domains. This may limit the generalizability of the findings across diverse linguistic and industrial environments. Practical Implication: The final focus of this paper is on the practical applications of speech recognition and text summarisation technologies in virtual assistants, meeting transcription, and content generation tools. This study benefits industry by improving operational efficiency, reducing manual effort, enabling real-time information processing, and enhancing productivity through AI-driven automation. Social Implication: Efficient transcription and information summarisation technologies help society manage the growing volume of digital information by improving accessibility, reducing information overload, supporting inclusive communication, and enabling faster access to relevant knowledge across education, healthcare, business, and public services. Originality/Value: The paper examines both speech recognition and text summarisation techniques, including deep learning and end-to-end models such as transformer networks for neural-network-based speech transcription, as well as advances in transformer-based summarisation models.
Zitieren
APA
Gode, C. S., Chati, P. U., Deshpande, P. P., Bhoyar, D. B., Tripathi, C. A., Kolte, J., Mohod, S. K., & Bhoyar, V. D. (2026). An Overview of Techniques for Speech-to-Text Conversion and Text Summarisation. AFRICAN JOURNAL OF APPLIED RESEARCH, 12(5), 897-916. https://doi.org/10.26437/b58e1n19
Vancouver
Gode CS, Chati PU, Deshpande PP, Bhoyar DB, Tripathi CA, Kolte J, et al. An Overview of Techniques for Speech-to-Text Conversion and Text Summarisation. AFRICAN JOURNAL OF APPLIED RESEARCH. 2026;12(5):897-916. doi:10.26437/b58e1n19
BibTeX
@article{Gode2026,
author = {Gode, C. S. and Chati, P. U. and Deshpande, P. P. and Bhoyar, D. B. and Tripathi, C .A. and Kolte, J. and Mohod, S. K. and Bhoyar, V. D.},
title = {An Overview of Techniques for Speech-to-Text Conversion and Text Summarisation},
journal = {AFRICAN JOURNAL OF APPLIED RESEARCH},
year = {2026},
volume = {12},
number = {5},
pages = {897--916},
doi = {10.26437/b58e1n19}
}
EUVIMED