EmoSpeech is a system capable of modulating the voice quality of a synthesizer while reading out aloud children’s tales, so that the voice conveys at least part of the emotions expressed by the corresponding text. This is achieved by controlling those parameters in the synthesizer that have been identified as having more relevance in the expression of emotions in human voice samples, based on manual evaluation by a set of volunteers. EmoSpeech operates with five basic emotions:anger, happiness, sadness, fear and surprise. The aspects of the voice that act as personality identifiers are: volume, rate, pitch baseline and pitch range. EmoTag uses a group of rules which relates the five basic emotions to the specific changes on voice parameters involved in the communication of emotion in human voice utterances.