This paper presents work using a new method of copy synthesis for the JSRU parallel formant synthesiser, to obtain good quality synthesis of both female and male speech. Natural speech was analysed to derive values for each of the synthesiser control parameters at regular (10 ms) intervals. The formant frequencies were obtained using excitation synchronous formant analysis and labelling techniques previously reported [10,11]. The amplitude control parameters were derived by a two-stage process: first the spectral amplitudes at the formant frequencies were measured from FFT analyses, and then the measured amplitudes were transformed using a table of "amplitude correction values" (ACVs) to obtain amplitudes which are appropriate as control parameters. Some modifications made to the synthesiser to improve the production of female speech are also described. The results have demonstrated that it is possible to obtain good quality female speech from the JSRU synthesiser.