Speech recognition synthesis systems for use in preschool period foreign language training

Developments in digital signal processing technology has led to the use of signal processing in various and far apart fields such as audio compression, development, and recognition synthesis. Computer science studies related to sound in general can be grouped under two main headings: speech recognition and speech synthesis. Speech syntesizer is a system that aims at carrying the text to an acoustic environment understandably and naturally. Similarly voiced expression recogniser is a system that aims at transforming the voice in acoustic environment to a text with acomputer. In recent years, the rapid improvements in the audio usual communication Technologies have made it inevitable to use these technologies in education. Foreign language teaching is one of the fields where the teacher and the students benefit from the education technologies. It? s our passage to adapt the improvements in voice technologies to foreign language education. Setting out with all these improvements, in this thesis sample, theinnovations in the are a of speech recognition and synthesis technology areaimed to be applied to the foreign language education. With this direction,an educational software in which these technologies of speech recognitionand combinements are integrated for the usage of the computer-supportedforeign language education has been prepared. With this software, it isplanned to contribute inviduals to acquire a language in early ages usingthe technology. In the light of this truth, the target audience has beendefined as pre-school children who are not acquainted with the ability toread and write. The design of the software is prepared as Microsoft Visual C# .NET 2005 programming language. The most distinct characteristic of thissoftware that differentiates itself from traditional foreign languagelearning softwares is due to its integrated recognition-synthesistechnology and its interface enriched with audio-visual elements, whichoffer the user both listening and pronounciation activities. The processesof reading of the words and transforming the words into text-picture in thesoftware have been proceeded under the control of Microsoft Speech Server.Statistic datas of the software evaluated according to expert views, havebeen calculated.