Recent Advances in End-to-End Automatic Speech Recognition
APSIPA Transactions on Signal and Information Processing · 20.04.2022 · 11 (1) · 1-64
DOI: 10.1561/116.00000050 ↗Kurz gefasst
Fachartikel in der Zeitschrift APSIPA Transactions on Signal and Information Processing, online erschienen am 20.04.2022. Verfasst von Jinyu Li. Die Arbeit wurde bisher 320-mal zitiert (Crossref) und nennt 351 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Recently, the speech community is seeing a significant trend of moving from deep neural network based hybrid modeling to end-to-end (E2E) modeling for automatic speech recognition (ASR). While E2E models achieve the state-of-the-art results in most benchmarks in terms of ASR accuracy, hybrid models are still used in a large proportion of commercial ASR systems at the current time. There are lots of practical factors that affect the production model deployment decision. Traditional hybrid models, being optimized for production for decades, are usually good at these factors. Without providing excellent solutions to all these factors, it is hard for E2E models to be widely commercialized. In this paper, we will overview the recent advances in E2E models, focusing on technologies addressing those challenges from the industry’s perspective.
Zitieren
APA
Li, J. (2022). Recent Advances in End-to-End Automatic Speech Recognition. APSIPA Transactions on Signal and Information Processing, 11(1), 1-64. https://doi.org/10.1561/116.00000050
Vancouver
Li J. Recent Advances in End-to-End Automatic Speech Recognition. APSIPA Transactions on Signal and Information Processing. 2022;11(1):1-64. doi:10.1561/116.00000050
BibTeX
@article{Li2022,
author = {Li, Jinyu},
title = {Recent Advances in End-to-End Automatic Speech Recognition},
journal = {APSIPA Transactions on Signal and Information Processing},
year = {2022},
volume = {11},
number = {1},
pages = {1--64},
doi = {10.1561/116.00000050}
}
EUVIMED