Manses: Turkish voice recognition and tr2en translation cloud

The purpose of Manses project is to design and develop a Turkish Speech Recognition (SR) and translation system, by using Cloud Computing technology. The base technology that will be used in this project is SR and as a final outcome of this project, it is expected to be achieved a Turkish SR system that is developed on the Cloud Computing architecture.

In mobile devices, processing and memory resources constraints don’t allow running a SR application on this devices. Therefore in this project it is aimed to give SR ability to mobile device using Cloud computing that has unlimited resource in order to SR.

The main purpose of the project is to help tourists to communicate with local population by using mobile devices. The outcome of the proposed project will translate speech into/from Turkish/English text format. The main scenario of the project is as follows: the speech of a Turkish speaker recorded by the system (via mobile phones or other mobile devices) will be translated into English and then will be presented to the user. The recorded sound file, with the aid of the mobile systems infrastructure, will be sent to the analysis center, where it will be analyzed by Cloud Computing technology.

As a touristic country, Turkey hosts thousands of tourists from various countries each year. Outcome of Manses project will make things possible for tourists who don’t know Turkish to acquire the English text version of a Turkish speech. Manses is going to make things possible for users who know Turkish and no any other languages when communicating with other people.

Speech recognition applications include voice user interfaces such as voice dialing (e.g., “Call home”), call routing (e.g., “I would like to make collect call”, search (e.g., find a podcast where particular words were spoken), simple data entry (e.g., entering a credit card number), preparation of structured documents (e.g., a radiology report), speech-to-text processing (e.g., word processors or emails), and aircraft (usually termed Direct Voice Input). In other hand, since speech processing technologies have a wide variety of application area among Turkic countries with population more than 200 millions of people, developing a robust recognition engine for Turkish language, accepted as the language of Silk Road, is a promising field in the next generation information systems consisting Turkish-specific speech agents.