Speech Recognition System
The first device for speech recognition came in 1952, it can recognize spoken man of numbers. In 1964, the Fair of computer technology in New York was represented by the unit ibm Shoebox. Commercial software speech recognition appeared in the early nineties. They are usually used by people who, because of hand injury is not able to recruit large amounts of text. These programs (eg, Dragon NaturallySpeaking, Voice Navigator) convert voice user text, thereby unloading his hands. Reliability of transfer of such programs is not very high, but over the years it has gradually improved.
Increasing computing power of mobile devices has allowed for them to create programs with speech recognition. Among such programs, it is worth noting the application Microsoft Voice Command, which allows you to work with many applications by voice. For example, you can include listening to the music player or a new document. Another exciting program is Speereo Voice Translator, voice translator. svt is able to recognize sentences spoken in English, and 'speak' in response to a translation into one of the chosen language. Intelligent Speech solutions that automatically identify and synthesize human speech, are the next step in the development of interactive voice systems (IVR). Using interactive phone application Currently, no trend, but a vital necessity.
Reducing the burden on operators of contact centers and secretaries, reducing labor costs and improve performance of service systems – these are just some advantages, demonstrating the feasibility of such decisions. Progress, however, does not stand still and, most recently in telephone interactive applications have been increasingly used in the automatic speech recognition and synthesis. In this case, communication with the voice portal is becoming more natural, as the choice it can be done not only through the touchtone, but with voice commands. At the same time recognition systems are independent from broadcasters, that is, recognize the voice of any person. The main advantage of voice systems is the user-friendliness – it gets rid of the need wade through complex and intricate maze of voice menus. Now it suffices to cast goal call, after which the voice system will automatically move the caller to a menu item. The next step of technology Speech recognition can be considered the development of so-called Silent Speech Interfaces (SSI) (Silent Access Interface). This speech processing system based on receipt and processing of speech signals in the early stages of articulating. This stage Development of speech recognition is caused by two major drawbacks of modern recognition systems: excessive sensitivity to noise, as well as the need for a concise and clear language when referring to a system of recognition. Approach, based on ssi, is to use the new sensors are not subject to the influence of noise in addition to processed acoustic signals.
