'Choral' speech has been defined by Tarnoczy as 'overlapped speech segments of equal duration obtained from a normal recording by the same speaker.' The analog technique used by Tarnoczy to produce choral speech consists of playing back simultaneously all the segments to be overlapped using several tape recorders, of which the outputs are connected in parallel to the input of another tape recorder. The computer assisted method described in this article is context independent, based on comparison of choral speech spectra. Results obtained from the reported experiment are similar to results obtained using other computer methods of VI/E, i.e., no errors of identification or elimination were observed when the samples involved were recorded through the same channel. The advantages of the choral methods are its independent-context characteristic and the need for less computer memory than the one required by other computer assisted methods of VI/E. 2 tables, 16 references.