Cutting tool condition monitoring is the key technique for realizing automatic and "un-manned" manufacturing processes. This project applies cutting force and acoustic emission transducers to monitor metal cutting processes. A B-spline neurofuzzy networks based tool wear state monitoring model has been presented. The model can accurately describe the nonlinear relation between the tool wear value and signal features. Compared with the normal neural networks, such as BP type ANNs, this model has the advantages of fast convergence and having local learning capabilities. Large amounts of monitoring experiments show that the application of B-spline neurofuzzy networks can improve the accuracy and reliability of the tool wear condition monitoring processes.