A new intelligent tool condition monitoring technique for metal cutting process is proposed. Fiest, the frequency spectrum analysis of the audio signal during cutting process, the pixel space projection analysis and equal gray pace analysis of the images of machined workpiece surface are introduced. Then combined the results of the audio signal analysis and the workpiece surface image analysis with artificial neural network, we implemented the intelligent tool condition monitoring based on multi-information fusion. The experimental results indicate that this method can recognize the tool condition effectively and improve the dependability of tool condition monitoring.