This paper reviews data on speech perception via implanted electrodes and via tactile aids. The two approaches are compared in terms of amount and types of aid provided to communication. Related issues discussed are the performance levels with multi-versus single-channel implants, promontory electrical stimulators versus implants, use of minimal residual hearing, implants for children, and the possible design of complementary systems combining auditory implant and tactile information. The diversity of the test methods and subjects used in implant versus tactile research precludes definitive comparisons of speech perception performance. However, it appears from the available data that, at present and for the foreseeable future, neither approach can provide more than a modest aid to lipreading. Speech reception test results from multichannel-implanted subjects are better, on the average, than for single-channel subjects. However, the best single-channel results are comparable to the best multichannel in tests using simple sentences. There is great variation among subjects with the same implant. Tactile aid performance by highly practiced subjects seems comparable to that of the better implant subjects.