Voice tracking error correction and feedback system that integrates CTC forced alignment
Data and Metadata · 09.05.2026 · 5 · 1467
DOI: 10.56294/dm20261467 ↗Kurz gefasst
Fachartikel in der Zeitschrift Data and Metadata, online erschienen am 09.05.2026. Verfasst von Tian Zhou. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Real-time speech follow-up reading systems require accurate phoneme-level error detection and timely corrective feedback, especially in children’s language training, foreign language learning, and speech assessment tasks.
Objective
The study aimed to develop a voice tracking error correction and feedback system integrating CTC forced alignment to improve pronunciation deviation localization, feedback timeliness, and multi-scenario speech interaction stability.
Methods
A CTC forced-alignment framework was combined with acoustic feature extraction, phoneme-level confidence modeling, dynamic feedback control, and edge-side deployment. The system was implemented on Jetson Xavier NX and evaluated using standard speech samples, error-prone repetition samples, and multi-speaker testing scenarios.
Results
The system achieved per-phoneme input latency below 290 ms and total feedback delay below 420 ms. Under moderate noise conditions, recognition accuracy reached 93,6 %, and feedback latency for common pronunciation errors remained below 210 ms. Across diverse speaker populations, the feedback success rate remained above 92,7 %.
Conclusions
The proposed system provided low-latency pronunciation error correction and stable feedback in follow-up reading tasks. Its containerized deployment and lightweight data transmission design supported cross-platform implementation, while future work should further validate multilingual adaptability and robustness under severe acoustic interference.
Zitieren
APA
Zhou, T. (2026). Voice tracking error correction and feedback system that integrates CTC forced alignment. Data and Metadata, 5, 1467. https://doi.org/10.56294/dm20261467
Vancouver
Zhou T. Voice tracking error correction and feedback system that integrates CTC forced alignment. Data and Metadata. 2026;5:1467. doi:10.56294/dm20261467
BibTeX
@article{Zhou2026,
author = {Zhou, Tian},
title = {Voice tracking error correction and feedback system that integrates CTC forced alignment},
journal = {Data and Metadata},
year = {2026},
volume = {5},
pages = {1467},
doi = {10.56294/dm20261467}
}
EUVIMED