A method for simplifying the control of a Klatt-type synthesizer is described in which values of a small set of high-level (HL) parameters are transformed into values for the larger set of low-level (LL) acoustic parameters that typify formant synthesizers. The HL parameters include natural frequencies of the vocal tract, area of oral constriction, specification of a nasal side branch, active control of intraoral pressure for obstruents, and state of the glottal source. Determination of the LL source parameters involves the calculation of a set of intermediate parameter values which reflect the pressures and flows within the vocal tract. To illustrate the method, we describe the synthesis of speechlike sounds in which the HL orifice parameters are varied through their ranges of values, as well as the synthesis of a number of consonants in medial (VCV) position.