Analysis of the correlation structure for a neural predictive model with application to speech recognition

By Li Deng, K. Hassanein, M. Elmasry. 1994

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Li Deng
2. Dong Yu
3. Jianfeng Gao
4. Xiaodong He
5. Yoshua Bengio
6. Zhe Gan
7. Mohit Bansal
8. Hamid Palangi

Citations in the corpus

• Deep Sentence Embedding Using Long Short-Term Memory Networks: Analysis and Application to Information Retrieval (2015, methodology): “The main idea of using RNN for sentence embedding is to find a dense and low dimensional semantic representation by sequentially and recurrently processing each word in a sentence and mapping it into a low dimensional vector.”
• Convolutional Neural Networks for Speech Recognition (2014, methodology): “They have been used in continuous speech recognition as feature extractors, in both the TANDEM approach [20], [21] and in so-called bottleneck feature methods [22], [23], [24], and also as nonlinear predictors to aid the recognition of speech units [25], [26].”
• Learning Input and Recurrent Weight Matrices in Echo State Networks (2013, background): “The recurrent Neural Network (RNN) belongs to a general type of deep neural networks that model time sequences and dynamical systems [1, 2, 3, 4, 5, 6].”
• Dynamic Speech Models: Theory, Algorithms, and Applications (2006, methodology): “, [99–101]).”
• A Primal-Dual Method for Training Recurrent Neural Networks Constrained by the Echo-State Property (2013, background): “The use of recurrent or temporally predictive forms of neural networks for speech recognition dated back to early 1990’s (Robinson, 1994; Deng et al., 1994), with relatively low accuracy.”
• Distributed Compressive Sensing: A Deep Learning Approach (2015, methodology): “It has been used extensively in time sequence modelling [30], [31], [32], [33], [34], [35], [36], [37], [38].”