Until now, a number of cutting tools and cutting methods have been developed for improving the machining processes, however they have some difficulties resulting from tool wear and failure. This study considers on-line monitoring of tool wear and failure in metal cutting processes. The proposed on-line monitoring system consists of two parts: the one is the on-line data acquisition system with sensors and a DSP board through LabView and a web server, and the other is the prediction of tool wear and failure using an ART2 neural network. The system is installed at an on-site machine tool to monitor high speed steel (HSS) tools for cutting titanium alloys. A number of experiments are carried out to demonstrate the effectiveness of the proposed system, and the results show that the proposed system can be applied to monitoring of the tool wear and failure.