SAIL LABS Technology today announced the availability of an Arabic Language Version of its Media Mining System. This will be the first product of its kind that enables users to search as well as summarise Arabic news broadcasts in real time.
In addition to existing features, such as Speaker Identification, Named Entity Detection, and Topic Detection, a new feature, known as Keyword Translation, has been introduced for the first time with the Arabic version. Users who have no knowledge of Arabic can use English in their search queries to track occurrences of query keywords within the Arabic broadcasts.
The Arabic version of the Media Mining System has a user interface in Arabic, which allows the user to type in Arabic characters for search retrieval. The search feature is multilingual, so the user can enter a query in English to search for specific content in an Arabic broadcast. The speech recognition module is equipped with a vocabulary of over 230,000 words. This feature, coupled with context-sensitive statistical models, contributes to high recognition accuracy of broadcast news despite variations in acoustical conditions.
“We are proud to announce the release of the Arabic Media Mining System because it proves our ability to not only add new languages to our existing portfolio but also produce improved functionality like keyword translation along with the new languages,” says Christoph Prinz, Vice President of Development at SAIL LABS Technology.
The Media Mining System takes audio input from a variety of sources to produce an indexed output in real time. This output can then be searched for relevant information. The system has various applications, ranging from Broadcast News Indexing and Media Asset Management to Keyword-triggered Notifications, and Interactive News Services. The Media Mining System can also be integrated into various systems, ranging from those handling storage and search-and-retrieval to those delivering audio content.
“For the first time, applications like personalised news services, media logging, multimedia clipping, and interactive TV services for the Arabic-speaking community can benefit from intelligent information extraction based on speech and language processing. To bridge cultural barriers, this new version of the Media Mining System gives access to multimedia documents in foreign languages by means of keyword translation,” says Dr Thomas Netousek, CEO of SAIL LABS Technology.
Share on Social Media