M. Akbacak, H. Franco, M. Frandsen, S. Hasan, H. Jameel, A. Kathol, S. Khadivi, X. Lei, A. Mandal, S. Mansour, K. Precoda, C. Richey, D. Vergyri, W. Wang, M. Yang, and J. Zheng, “Recent advances in SRI’s IraqComm® Iraqi Arabic-English speech-to-speech translation system,” in Proc. 2009 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2009), pp. 4809–4813.
Abstract
We summarize recent progress on SRI’s IraqComm™ IraqiArabic-English two-way speech-to-speech translation system. In the past year we made substantial developments in our speech recognition and machine translation technology, leading to significant improvements in both accuracy and speed of the IraqComm system. On the 2008 NIST-evaluation dataset our two-way speech-to-text (S2T) system achieved 6% to 8% absolute improvement in BLEU in both directions, compared to our previous year system [1].