Converting text to speech or the so called "speech synthesis" is the artificial production of human speech. A text to speech system converts normal language text to speech. The two most important factors of the process are the similarity of the artificial voice to a human one and its ability to be understood.



Nowadays computers are used to do this kind of conversion. The first computer that was able to do this was developed in the late 1950s and the first complete text-to-speech system was completed in the early 70s. Since then the process of developing better systems continues. The main challenges in front the people developing TTS systems are to improve the semantic representation of the text being converted, to improve the pronunciation according the punctuation within the text, to improve the pronunciation of the words within the text as well as the numbers.

The most common use of speech synthesis systems is providing accessibility for people with serious disabilities. Visually impaired, people with dyslexia or other reading difficulties can use computers, read and work normally because of such systems. TTS systems are also used by pre-literate children, by companies seeking ways to improve their customer service or developing mobile phone applications.

It is not hard to get a text synthesis converter. Nowadays all computer operation systems support it. Apple was the first that integrated a TTS system into their operation system. During the years they invested into developing the feature and now their operation systems come with a fully supported program to text conversion Plain Talk. Windows users can also benefit from TTS integrated into the operating system. Windows systems also feature a voice recognition engine so that voice commands can be given to the computer. All Windows versions after Windows 95 have a speech synthesis program called Narrator, directly available to users. Moreover Windows Server 2003 features a complete package for voice synthesis and recognition, for commercial applications such as call centers. Linux users can also benefit from text to speech applications such as Festival Speech Synthesis System.

Internet also offers various text to speech synthesis tools. Power Text to Speech Reader and Text Aloud are just an example of software that can read messages from a mail client or web browser. Since the latest trend in web development is mainly to make web more accessible, many websites implement narrators to help disabled people read their websites.