Artificial intelligence-based speech recognition system

Speech recognition becomes popular for information technologies in recent years. Translating words to the digital media is being easier thanks to the speech recognition. Most of the electronic devices can be controlled by voices using speech recognition technologies.In this study, basics methods for speech recognition are described briefly, the most popular artificial intelligence methods NN and ANFIS was also described. At first in the realized systems the features vectors of speech signals based on wavelet transform entropy and mel frequency cepstral coefficients are evaluated. After than these feature vectors are used for training of artificial intelligence based speech recognizer models. The trained models are tested for speech determination. Obtained results are tabulated.Keywords: Speech recognition, neural network, Adaptive Network Based Fuzzy Inference System, Entropy, Wavelet Transform