For many manufacturers, Text-To-Speech (TTS) systems are available in only one (e.g. Hungarian) language or at most two (one male and one female) voices. Practical applications are hampered if several applications in a given language use the same voice. An example is the case of people with speech impairments: it would be strange if all patients in a hospital used the same voice. The project will implement a complex server-based application that will allow the creation of personalised voice recordings, mainly in Hungarian, in a quality-assured way, which will later allow the generation (even automatically) of a TTS engine that will speak the given voice. Both server and client side work is required. The task can be continued up to thesis/diploma/TDK level, good preparation for PhD studies.

For many manufacturers, Text-To-Speech (TTS) systems are available in only one (e.g. Hungarian) language or at most two (one male and one female) voices. Practical applications are hampered if several applications in a given language use the same voice. An example is the case of people with speech impairments: it would be strange if all patients in a hospital used the same voice. The project will implement a complex server-based application that will allow the creation of personalised voice recordings, mainly in Hungarian, in a quality-assured way, which will later allow the generation (even automatically) of a TTS engine that will speak the given voice. Both server and client side work is required. The task can be continued up to thesis/diploma/TDK level, good preparation for PhD studies.