Modular Construction of Time-Delay Neural Networks for Speech Recognition

By A. Waibel. 1989

How researchers see this paper

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

Citations in the corpus

Time-series forecasting with deep learning: a survey (2020, methodology): “Furthermore, older models of neural networks have been used historically in time-series applications, as seen in [24] and [25].”
The Tempo 2 Algorithm: Adjusting Time-Delays By Supervised Learning (1990, methodology): “The TDNN has been shown to be a very powerful approach to phoneme recognition.”
Policy-Gradient Algorithms for Partially Observable Markov Decision Processes (2003, methodology): “Good results were obtained for classifying plosive consonants using TDNNs compared to standard ANNs [Waibel, 1989].”
Sequence learning with recurrent networks: analysis of internal representations (1992, methodology): “Such explicit representations of time have been used in NETtalk [11] and \time-delay neural networks" such as those used with some success in speech recognition [13].”
Continual learning in reinforcement environments (1995, methodology): “A more sophisticated method for dealing with delay lines is the TDNN (Time-Delay Neural Network) described by Waibel [110].”
A time delay neural network architecture for efficient modeling of long temporal contexts (2015, methodology): “This architecture uses a modular and incremental design to create larger networks from sub-components [3].”
Zero-shot Learning with Semantic Output Codes (2009, background): “To achieve vocabulary independence, speech recognition systems typically employ a phoneme-based recognition strategy (Waibel, 1989).”
Learning the 2-D Topology of Images (2007, background): “On the other hand, other learning algorithms successfully exploit the specific topology of their input, e.g., SIFT-based machine vision [10], convolutional neural networks [6, 7], time-delay neural networks [5, 16].”
Time series prediction by using a connectionist network with internal delay lines (1993, methodology): “1 Alternative Representations of the FIR Topology A similar network structure incorporating embedded time delays is the Time-Delay Neural Network [13, 14, 22].”
Learning to Forget: Continual Prediction with LSTM (2000, unknown): “Like standard LSTM, extended LSTM is local in space and time.”