Morse code, with its single switch operation, has been shown to be an excellent adaptive communication device for the disabled, especially those with impaired hand coordination and dexterity. Because a stable typing rate is required for an accurate recognition of Morse code, it presents a major hindrance to disabled persons in applying Morse code as a useful communication tool. Therefore, a suitable adaptive recognition method is needed. This paper applies the Least-Mean- Square algorithm to an adaptive Chinese phonetic Morse code recognition system which includes four processes: space recognition, tone recognition, adaptive processing, and character recognition. Statistical analyses demonstrated that the proposed method results in a better recognition rate compared to alternative methods in the literature.