Convolutional Neural Networks (CNNs) are effective models for reducing spectral variations and modeling spectral correlations in acoustic features for automatic speech recognition (ASR). Hybrid speech recognition systems incorporating CNNs with Hidden Markov Models/Gaussian Mixture Models (HMMs/GMMs) have achieved the state-of-the-art in various benchmarks. Meanwhile, Connectionist Temporal Classification (CTC) with Recurrent Neural Networks (RNNs), which is proposed for labeling unsegmented sequences, makes it feasible to train an 'end-to-end' speech recognition system instead of hybrid settings. However, RNNs are computationally expensive and sometimes difficult to train. In this paper, inspired by the advantages of both CNNs and the CTC approach, we propose an end-to-end speech framework for sequence labeling, by combining hierarchical CNNs with CTC directly without recurrent connections. By evaluating the approach on the TIMIT phoneme recognition task, we show that the proposed model is not only computationally efficient, but also competitive with the existing baseline systems. Moreover, we argue that CNNs have the capability to model temporal correlations with appropriate context information.

No abstract

Nowadays customers would rather buy their needs online than visiting a retail store because of many reasons such as saving time. Therefore, in order to increase efficiency of online shopping websites, many companies have invested in researches toward prediction of users purchases and recommendation sys tems that may help and motivate a user to buy products that he may be interested in. However, most efforts in this area has been around classification and predictions based on users interests in specific types of products. In this paper, we have studied efficiency of numerous algorithms toward building a classification model to predict the probability of a complete purchase by users only based on their behavior models in the system and regardless of their interest. Therefore, we experimented accuracy of difl'erent algorithms and proposed a novel classification model that is able to predict whether a user will be interested in buying a certain set of products that are placed in the online shopping cart or not.

No abstract