DocumentCode :
3427572
Title :
Parsing-based objective functions for speech recognition in translation applications
Author :
Hillard, D. ; Hwang, M. ; Harper, M. ; Ostendorf, M.
Author_Institution :
Dept. of Electr. Eng., Washington Univ., Seattle, WA
fYear :
2008
fDate :
March 31 2008-April 4 2008
Firstpage :
5109
Lastpage :
5112
Abstract :
This paper looks at a parsing-based alternative to word error rate (WER) for optimizing recognition, SParseval, hypothesizing that it may be a better objective for applications such as translation. We find that SParseval is more correlated than WER with human measures of subsequent translation performance, but that optimizing explicitly for SParseval does not give a significant reduction in translation error as measured by automatic methods based on a single translation reference. However, anecdotal examples indicate that SParseval does improve automatic speech recognition (ASR) results, leaving open the possibility that it may be more useful in the future or for other language processing tasks.
Keywords :
error statistics; grammars; language translation; speech processing; speech recognition; SParseval; automatic speech recognition; machine translation; parsing-based objective functions; speech recognition; word error rate; Application software; Automatic speech recognition; Computer science; Data mining; Educational institutions; Error analysis; Gold; Natural languages; Performance gain; Speech recognition; parsing; speech recognition objective; speech translation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on
Conference_Location :
Las Vegas, NV
ISSN :
1520-6149
Print_ISBN :
978-1-4244-1483-3
Electronic_ISBN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2008.4518808
Filename :
4518808
Link To Document :
بازگشت