Paper Info
Title
Let-Decoder: A Wfst-Based Lazy-Evaluation Token-Group Decoder With Exact Lattice Generation
Abstract
We propose a novel lazy-evaluation token-group decoding algorithm with on-the-fly composition of weighted finite-state transducers (WFSTs) for large vocabulary continuous speech recognition. In the standard on-the-fly composition decoder, a base WFST and one or more incremental WFSTs are composed during decoding, and then token passing algorithm is employed to generate the lattice on the composed search space, resulting in substantial computation overhead. To improve speed, the proposed algorithm adopts 1) a token-group method, which groups tokens with the same state in the base WFST on each frame and limits the capacity of the group and 2) a lazy-evaluation method, which does not expand a token group and its source token groups until it processes a word label during decoding. Experiments show that the proposed decoder works notably up to 3 times faster than the standard on-the-fly composition decoder.
Year
DOI
Venue
2021
10.1109/LSP.2021.3067220
IEEE SIGNAL PROCESSING LETTERS
Keywords
DocType
Volume
Decoding, Lattices, Speech recognition, Histograms, Signal processing algorithms, Hidden Markov models, Standards, On-the-fly composition, on-the-fly lattice rescoring, speech recognition, WFST
Journal
28
ISSN
Citations 
PageRank 
1070-9908
0
0.34
References 
Authors
0
7
Name
Order
Citations
PageRank
Hang Lv102.70
Daniel Povey22442231.75
Mahsa Yarmohammadi391.23
Ke Li45026.41
Yiming Wang5173.27
Lei Xie65211.82
Sanjeev Khudanpur72155202.00