Spyros Matsoukas
Research Subjects
Spyros Matsoukas has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
Bing Xiang
BBN Technologies, 10 Moulton St., Cambridge, MA 02138
1 shared publication- Hamdy Mubarak
Arabic NLP Researche and Development Department, Sakhr Software Co
1 shared publication - Hazem Nader
Sakhr Software, Free Zone
1 shared publication Ivan Bulyko
BBN Technologies, 10 Moulton St., Cambridge, MA 02138
1 shared publicationJohn Makhoul
BBN Technologies, 10 Moulton St., Cambridge, MA 02138
1 shared publication- John Makhoul
Raytheon BBN Technologies Microsoft Research 10 Moulton St One Microsoft Way 02138 98052 Cambridge Redmond MA WA USA
1 shared publication
Latest Publications
- 22 September 2011
- Published by Springer Nature in Machine Translation
- Vol. 26 (1-2) , 67-83
Methods for integrating rule-based and statistical systems for Arabic to English machine translation
- 22 September 2011
- Published by Springer Nature in Machine Translation
- Vol. 26 (1-2) , 67-83
Methods for integrating rule-based and statistical systems for Arabic to English machine translation
- 1 April 2007
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 4 (15206149) , IV-1281-1281
Integrating Speech Recognition and Machine Translation
This paper presents a set of experiments that we conducted in order to optimize the performance of an Arabic/English machine translation system on broadcast news and conversational speech data. Proper integration of speech-to-text (STT) and machine translation (MT) requires special attention to issues such as sentence boundary detection, punctuation, STT accuracy, tokenization, conversion of spoken numbers and dates to written form, optimization of MT decoding weights, and scoring. We discuss these issues, and show that a carefully tuned STT/MT integration can lead to significant translation accuracy improvements compared to simply feeding the regular STT output to a text MT system.
- 1 April 2007
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 4 (15206149) , IV-1281-1281
Integrating Speech Recognition and Machine Translation
This paper presents a set of experiments that we conducted in order to optimize the performance of an Arabic/English machine translation system on broadcast news and conversational speech data. Proper integration of speech-to-text (STT) and machine translation (MT) requires special attention to issues such as sentence boundary detection, punctuation, STT accuracy, tokenization, conversion of spoken numbers and dates to written form, optimization of MT decoding weights, and scoring. We discuss these issues, and show that a carefully tuned STT/MT integration can lead to significant translation accuracy improvements compared to simply feeding the regular STT output to a text MT system.