Framewise phoneme classification with bidirectional LSTM networks

By Alex Graves, J. Urgen Schmidhuber. Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005., 2005

710 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

View Paper: https://www.semanticscholar.org/paper/685d42a668413422615519a52ac75d66fded4611

How researchers see this paper

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

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Machine Learning for Wireless Networks with Artificial Intelligence: A Tutorial on Neural Networks (2017, methodology): “The output of the current time step is then generated by combining both layers’ hidden vector [137].”
RNN-based Learning of Compact Maps for Efficient Robot Localization (2007, methodology): “In order to maximize the temporal context available for localization, a bidirectional Long-Short Term Memory architecture is used [10, 11] for the neural ∗This work was supported by the Sixth Framework Program of the European Union (project number IST-511931) and by the Swiss National Science…”
Minimal gated unit for recurrent neural networks (2016, methodology): “LSTM learning follows typical stochastic gradient descent and back propagation (Graves & Schmidhuber, 2005).”
Controllable Emphasis with zero data for text-to-speech (2023, methodology): “It consists of a TACOTRON2 [24] phomeme encoder, a phonemeto-frame upsampler which is smoothed with a Bi(directional)LSTM [25].”
Video Summarization with Long Short-Term Memory (2016, methodology): “We use bidirectional LSTM layers [44] for modeling better long-range dependency in both the past and the future directions.”
A Review of Deep Learning Based Speech Synthesis (2019, methodology): “To address these problems, the authors in [46] proposed a modeling method based on recurrent neural networks (RNNs).”
Classifying Unprompted Speech by Retraining LSTM Nets (2005, methodology): “Current results with BLSTM [9] show that it outperforms normal LSTM, as well as support vector machines and previous bidirectional RNNs on speech recognition tasks.”
Understanding LSTM - a tutorial into Long Short-Term Memory Recurrent Neural Networks (2019, neutral): “In the early, ground-breaking papers by Hochreiter [41] and Graves [34], the authors used different notations which made further development prone to errors and inconvenient to follow.”
Neural Architectures for Named Entity Recognition (2016, methodology): “…networks using the back-propagation algorithm updating our parameters on every training example, one at a time, using stochastic gradient descent (SGD) with 2(Graff, 2011; Biemann et al., 2007; Callison-Burch et al., 2010; Parker et al., 2009) a learning rate of 0.01 and a gradient clipping of 5.0.”
Sentiment Analysis using Imperfect Views from Spoken Language and Acoustic Modalities (2018, methodology): “Sequence labeling with LSTM-RNNs can account for contextual information from adjacent inputs as well as the overall input sequence and has been shown to perform better on several tasks (Graves and Schmidhuber, 2005; Graves et al., 2008; Poria et al., 2017; Sheikh et al., 2017).”
Exploring architectures, data and units for streaming end-to-end speech recognition with RNN-transducer (2017, background): “Deep recurrent neural networks with long short-term memory (LSTM) cells [2] have recently been shown to be ideal for this task [3, 4, 5].”
Temporal Models for Predicting Student Dropout in Massive Open Online Courses (2015, background): “Over the past decade, LSTM has been applied to various real-world problems, such as protein structure prediction [24], speech recognition [25], [26] and handwriting recognition [27], [28].”
A review on the long short-term memory model (2020, methodology): “Graves and Schmidhuber (2005b) presented the BLSTM network, comparing it with the unidirectional variant in a classiﬁcation context.”
Efficient Epileptic Seizure Prediction Based on Deep Learning (2019, methodology): “Instead of using LSTM as the classifier, we used a Bidirectional-LSTM (Bi-LSTM) network [28] in which each”
MOrdReD: Memory-based Ordinal Regression Deep Neural Networks for Time Series Forecasting (2018, methodology): “This bidirectional model has shown remarkable success in the NLP literature [38], including extensions to LSTM-based models such as the sequence-to-sequence architecture [19].”
Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network (2018, background): “In contrast, the present analysis follows the method described in [15] and [62], [61], [9], [10].”