Abstract | ||
---|---|---|
We describe the large vocabulary automatic speech recognition system developed for Modern Standard Arabic by the SRI/Nightingale team, and used for the 2007 GALE evaluation as part of the speech translation system. We show how system performance is affected by different development choices, ranging from text processing and lexicon to decoding system architecture design. Word error rate results are reported on broadcast news and conversational data from the GALE development and evaluation test sets. Index Terms: speech recognition, large vocabulary, Arabic |
Year | Venue | Keywords |
---|---|---|
2008 | INTERSPEECH | automatic speech recognition,system performance,system architecture,word error rate,speech recognition,indexing terms |
Field | DocType | Citations |
Speech analytics,Computer science,Word error rate,Speech recognition,Lexicon,Modern Standard Arabic,Natural language processing,Artificial intelligence,Speech translation,Vocabulary,Speech technology,Text processing | Conference | 14 |
PageRank | References | Authors |
2.68 | 10 | 12 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dimitra Vergyri | 1 | 373 | 36.97 |
Arindam Mandal | 2 | 158 | 16.44 |
Wen Wang | 3 | 327 | 29.31 |
Andreas Stolcke | 4 | 6690 | 712.46 |
Jing Zheng | 5 | 442 | 43.00 |
Martin Graciarena | 6 | 281 | 24.70 |
David Rybach | 7 | 188 | 20.31 |
Christian Gollan | 8 | 260 | 18.73 |
Ralf Schlüter | 9 | 1337 | 136.18 |
Katrin Kirchhoff | 10 | 1026 | 95.24 |
Arlo Faria | 11 | 66 | 7.87 |
Nelson Morgan | 12 | 3048 | 533.52 |