DocumentCode
3514911
Title
Systems combination in large vocabulary continuous speech recognition
Author
Abida, Kacem ; Karray, Fakhri
Author_Institution
Electr. & Comput. Eng. Dept., Univ. of Waterloo, Waterloo, ON, Canada
fYear
2010
fDate
21-23 June 2010
Firstpage
1
Lastpage
6
Abstract
We present a review of the most significant advances in the field of system combination towards reducing word error rates (WER) in large vocabulary continuous speech recognition (LVCSR). We have mainly focused on Recognizer Output Voting Error Reduction (ROVER), confusion network (CN) and minimum frame word error rate (fWER) based combination along with the latest improvements. Despite lot of progress witnessed in this field, some challenges still remain in enhancing the performance of LVCSR. We suggest in this paper some directions that may lead to a lower WER within the framework of system combination.
Keywords
speech recognition; vocabulary; confusion network; frame word error rate; large vocabulary continuous speech recognition; recognizer output voting error reduction; systems combination; Computer numerical control; Context; Decoding; Lattices; Semantics; Speech; Speech recognition; ASR; Confusion Network; ROVER; System Combination; WER; fWER decoding;
fLanguage
English
Publisher
ieee
Conference_Titel
Autonomous and Intelligent Systems (AIS), 2010 International Conference on
Conference_Location
Povoa de Varzim
Print_ISBN
978-1-4244-7104-1
Type
conf
DOI
10.1109/AIS.2010.5547038
Filename
5547038
Link To Document