Adaptive Phoneme State Learning Architecture for Enhanced Speech Recognition Using Backpropagation Neural Network and Hidden Markov Model
F1000Research · 2026
DOI: 10.12688/f1000research.177414.2 ↗Kurz gefasst
Fachartikel in der Zeitschrift F1000Research, 2026. Verfasst von Rashmi Siddalingappa und 8 weiteren Autor:innen. Beteiligte Einrichtung: Computer and Data Science, York St John University, London, England, E14 2BA, UK. Hauptthemen laut MEDLINE: Neural Networks, Computer, Speech, Speech Recognition Software. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Speech remains a primary mode of human communication; however, automated speech recognition (ASR) systems face challenges from accent variability, temporal fluctuations, noise, and data privacy concerns. This paper proposes an enhanced ASR architecture incorporating an Adaptive Phoneme State Learning (APSL) algorithm with a Backpropagation Neural Network (BPNN) and Hidden Markov Model (HMM). APSL dynamically adjusts HMM state probabilities using phoneme confidence scores derived from the BPNN, thereby improving phoneme transition modeling and alignment. The multi-stage ASR pipeline includes noise reduction, speech-pause detection, and feature extraction via framing and windowing. APSL’s adaptive mechanism reduces ambiguities in phoneme transitions, resulting in a more accurate speech-to-text conversion. A comparative evaluation framework assesses the baseline HMM, standalone BPNN, and integrated APSL-BPNN-HMM model. Experiments were conducted using a custom-built dataset of 2000 audio files alongside five benchmark corpora: BNC, ANC, COCA, Buckeye, and Emu. Key evaluation metrics—recall, precision, F-score, and Word Error Rate (WER)—demonstrate that the APSL-enhanced model significantly outperforms baseline systems, achieving 95.7% recall, 92.95% precision, 94.53% F-score, and 96% overall accuracy. Notably, APSL-BPNN-HMM consistently yielded the lowest WER across all datasets, validating its effectiveness. This work highlights the benefits of adaptive learning in probabilistic frameworks for achieving robust and accurate speech recognition.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Computer and Data Science, York St John University, London, England, E14 2BA, UK.
- Christ University, Bengaluru, Karnataka, India.
- Department of Computer Science, Rani Channamma University, Belagavi, Karnataka, India.
- UST Global, Bangalore, Karnatake, India.
- Macao Polytechnic University, Macau, Macao.
- INTI International University & Colleges, Nilai, Negeri Sembilan, Malaysia.
ORCID-Profile: Rashmi Siddalingappa, Shivanand Gornale
Zitieren
APA
Siddalingappa, R., S, D., Savitha, M., P, K., Stella Mary I, P., Gornale, S., B A, L., Li, K., & Wen Goh, K. (2026). Adaptive Phoneme State Learning Architecture for Enhanced Speech Recognition Using Backpropagation Neural Network and Hidden Markov Model. F1000Research. https://doi.org/10.12688/f1000research.177414.2
Vancouver
Siddalingappa R, S D, Savitha M, P K, Stella Mary I P, Gornale S, et al. Adaptive Phoneme State Learning Architecture for Enhanced Speech Recognition Using Backpropagation Neural Network and Hidden Markov Model. F1000Research. 2026. doi:10.12688/f1000research.177414.2
BibTeX
@article{Siddalingappa2026,
author = {Siddalingappa, Rashmi and S, Deepa and Savitha, Margaret and P, Kalpana and Stella Mary I, Priya and Gornale, Shivanand and B A, Lakshmi and Li, Kefeng and Wen Goh, Khang},
title = {Adaptive Phoneme State Learning Architecture for Enhanced Speech Recognition Using Backpropagation Neural Network and Hidden Markov Model},
journal = {F1000Research},
year = {2026},
doi = {10.12688/f1000research.177414.2}
}
EUVIMED