SOURCE: Applications Technology, Inc (AppTek)

Applications Technology, Inc (AppTek)

June 22, 2010 06:45 ET

AppTek Releases Arabic, Farsi, Pashto and Urdu Language Engines for Its Hybrid Machine Translation System

MCLEAN, VA--(Marketwire - June 22, 2010) -  AppTek, a leader in human language technology (HLT), which was the first company to release a true Hybrid Machine Translation (HMT) platform, announced the release of Arabic, Farsi, Pashto, and Urdu for its patent-pending hybrid machine translation system. AppTek's hybrid machine translation system is a full integration of statistical and rule-based methodologies. This system has been ranked first among commercial machine translation (MT) systems tested during an independent comparative study. AppTek's HMT system was ranked first overall for in processing noisy data 3 consecutive years by the National Institute of Standards and Technology (NIST).

"AppTek is the only human language technology provider to successfully develop a single system to enable fully-automatic, high quality machine translation, to integrate rule-based and statistical features," said Hassan Sawaf, Chief Scientist at AppTek. "The need to release Arabic, Farsi, Pashto, and Urdu on our Hybrid Machine Translation system has been driven by the increasing demand for translation of speech and text data in these languages in recent years. Moreover, we wanted to provide clients with our unique ability to be highly accurate in the process of automated translation of large volumes of data, and provide greater information fidelity in these languages."

AppTek's HMT solution provides a full integration of both MT methodologies, rather than simply adding rules to the statistical system or a minor statistical module to the rule-based engine. Using the company's statistical MT platform and augmenting it with its rich rule-based MT engine, AppTek's HMT solution is pushing the state-of-the-art in MT design to the next level. The three key translation quality parameters of MT systems -- fluency, informativeness, and adequacy -- are now supported in one comprehensive system with greater performance, quality and accuracy.

For a quick trial of AppTek's Hybrid Machine Translation please visit and try Quick Translate. For more extensive translations such as word documents and web page translation request a free WebTrans login through AppTek's demo request form at:


AppTek, headquartered in McLean, Virginia, is a developer of human language technology products with a complete suite for text and speech (voice) processing and recognition. The company also leads major research and development efforts to further the advancement in the field of developing better methods and technologies in the field of HLT. AppTek's product offerings include machine translation (MT) and automatic speech recognition (ASR) for a growing list of more than 23 languages; multilingual information retrieval with query and topic search capabilities; name-finding applications; and integrated suites providing ASR and MT in media monitoring of broadcast and telephony speech as well as handheld and wearable speech-to-speech translation devices. The company has language professionals and computer scientists in offices around the world.

Contact Information