Ladan Khosravani Pour - Department of Electrical Engineering, South Tehran Branch, Islamic Azad University, Tehran, Iran
Ali Farrokhi - Department of Electrical Engineering, South Tehran Branch, Islamic Azad University, Tehran, Iran

Speech recognition representing a communication between computers and human as a sub field of computational linguistics or natural language processing has a long history. Automatic Speech Recognition (ASR), Text to Speech (TTS), speech to text, Continuous Speech Recognition (CSR), and interactive voice response systems are different approaches to solving problems in this area. The performance improvement is partially attributed to the ability of the Deep Neural Network (DNN) to model complex correlations in speech features. In this paper, unlike the use of conventional model for sequential data like voice that employs Recurrent Neural Network (RNNs) with the emergence of different architectures in deep networks and good performance of Conventional Neural Networks (CNNs) in image processing and feature extraction, the application of CNNs was developed in other domains. It was shown that prosodic features for Persian language could be extracted via CNNs for segmentation and labeling speech for short texts. By using ۱۲۸ and ۲۰۰ filters for CNN and special architectures, ۱۹.۴۶ error in detection rate and better time consumption than RNNs were obtained. In addition, CNN simplifies the learning procedure. Experimental results show that CNN networks can be a good feature extractor for speech recognition in various languages.

Speech recognition, Speech Segmentation, convolutional neural networks, Persian Language, artificial intelligence