BACKGROUND: Studies on computer assisted therapy (CAT) show encouraging outcomes with chronic aphasic speakers. However, there have been no studies investigating efficacy of CAT carried out with acute in-patients. AIMS: This paper aims to evaluate the effects and feasibility of an unsupervised computer-based therapy for anomia in chronic out-patient and acute in-patient aphasic participants. The computerised training programmes were selected according to each participant's anomic syndrome. METHODS AND PROCEDURES: A multiple single-case design was used with two subgroups of aphasic participants. In the out-patients group (N = 4), CAT sessions were alternated with an equal number of clinical treatment sessions. In a second group of seven in-patients with acute aphasia, CAT was added to daily individual aphasia therapy. OUTCOMES AND RESULTS: In four chronic out-patients, a significant item-specific effect of CAT was shown. For two participants, this effect was similar to the effect of an equal number of individual clinical therapy sessions. Results were more heterogeneous in the in-patient group: an item-specific effect of CAT, not accounted for by spontaneous recovery, was observed in three of the seven participants. CONCLUSIONS: The present data suggest that individually adapted CAT can be effective as an adjunct to clinical therapy for anomia, not only with chronic aphasic out-patients but also in acute in-patients. Further investigation is needed in order to specify the conditions of application of CAT, given the varied results among our participants specially in the in-patients group.