At present, users of text-to-speech products can select either male or female voice, or even a multitude of different languages for various applications. However, the next generation of text-to-speech converters will need to provide additional features, such as the capability of switching from one dialect of a language to another. This paper shows that if a given language, other dialects can easily be derived from the base dialect. This paper will provide a strategy for synthesizing examples of different dialects of English and Spanish.