Publication Details
Out-of-Vocabulary Word Recovery Using FST-Based Subword Unit Clustering in a Hybrid ASR System
Out-of-vocabulary Words, Robust ASR
The paper presents a new approach to extracting useful information from out-of-vocabulary (OOV) speech regions in ASR system output. The system makes use of a hybrid decoding network with both words and sub-word units. In the decoded lattices, candidates for OOV regions are identified as sub-graphs of sub-word units. To facilitate OOV word recovery, we search for recurring OOVs by clustering the detected candidate OOVs. The metrics for clustering is based on a comparison of the sub-graphs corresponding to the OOV candidates. The proposed method discovers repeating outof- vocabulary words and finds their graphemic representation more robustly than more conventional techniques taking into account only one best sub-word string hypotheses.
@INPROCEEDINGS{FITPUB11725,
author = "Ekaterina Egorova and Luk\'{a}\v{s} Burget",
title = "Out-of-Vocabulary Word Recovery Using FST-Based Subword Unit Clustering in a Hybrid ASR System",
pages = "5919--5923",
booktitle = "Proceedings of ICASSP 2018",
year = 2018,
location = "Calgary, CA",
publisher = "IEEE Signal Processing Society",
ISBN = "978-1-5386-4658-8",
doi = "10.1109/ICASSP.2018.8462221",
language = "english",
url = "https://www.fit.vut.cz/research/publication/11725"
}