Abstract: In this study, we present a speech recognition interface designed for vocal control. The implementation has been realized under the Matlab environment with scripts in C. The program uses the statistical HMM (Hidden Markov Models) for speech modeling, the K-means, Baum-welch algorithms for training and codebook conception and finally the Viterbi decoding algorithm for the recognition process. The recognized word decision is based on the maximal likehood value. The speech database is constituted of 1000 words mono-speaker associated with a denoising module before be applied to the developed interface.