Performative time and pitch scaling is a new research paradigm for prosodic analysis by synthesis. In this paper, a system for real-time recorded speech time and pitch scaling by the means of hands or feet gestures is designed and evaluated. Pitch is controlled with the preferred hand, using a stylus on a graphic tablet. Time is controlled using rhythmic frames, or constriction gestures, defined by pairs of control points. The " Arsis " corresponds to the constriction (weak beat of the syllable) and the " Thesis " corresponds to the vocalic nucleus (strong beat of the syllable). This biphasic control of rhythmic units is performed by the non-preferred hand using a button. Pitch and time scales are modified according to these gestural controls with the help of a real-time pitch synchronous overlap-add technique (RT-PSOLA). Rhythm and pitch control accuracy are assessed in a prosodic imitation experiment: the task is to reproduce intonation and rhythm of various sentences. The results show that inter-vocalic durations differ on average of only 20 ms. The system appears as a new and effective tool for perfor-mative speech and singing synthesis. Consequences and applications in speech prosody research are discussed.