Robust speech recognition using long short-term memory recurrent neural networks for hybrid acoustic modelling

By Jürgen T. Geiger, Zixing Zhang, Felix Weninger, Björn Schuller, Gerhard Rigoll. 2014

75 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

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

Deep learning in neural networks: An overview (2014, background): “, 2014), medium-vocabulary speech recognition (Geiger et al., 2014), prosody contour prediction (Fernandez et al.”
RUDDER: Return Decomposition for Delayed Rewards (2018, background): “and language applications, including handwriting recognition [37], generation of writings [38], language modeling and identiﬁcation [35, 146], automatic language translation [126], speech recognition [107, 29] analysis of audio data [78], analysis, annotation, and description of video data [22, 136,”
Explaining and Interpreting LSTMs (2019, background): “ch and language applications, including handwriting recognition [24], generation of writings [23], language modeling and identication [22,78], automatic language translation [73], speech recognition [63,17], analysis of audio data [49], analysis, annotation, and description of video data [15,76,70].”
Deep Learning for Environmentally Robust Speech Recognition (2017, methodology): “Althrough some supervised machine-learning approaches have been proposed, such as Non-negative Matrix Factorisation (NMF) (Geiger et al. 2014a; Lee and Seung 1999; Schuller et al. 2010; Weninger et al. 2012), they struggle to obtain effective representations (aka dictionaries) of noise and speech in”
Modeling speaker variability using long short-term memory networks for speech recognition (2015, methodology): “LSTM RNNs have been successfully used for many sequence labeling and sequence prediction tasks, such as hand-writing recognition [18], speech recognition [19][28][29], speech synthesis [20], and language identiﬁcation (LID) [21].”
Constructing long short-term memory based deep recurrent neural networks for large vocabulary speech recognition (2014, background): “Recen tly, LSTM networks have also been introduced on phoneme recognit i n task [6], robust speech recognition task [14], and large voc abulary speech recognition task [15][16][17], and shown state-ofthe-art performances.”
Speech enhancement and recognition using multi-task learning of long short-term memory recurrent neural networks (2015, methodology): “In [8], the LSTM was applied in a medium-vocabulary scenario, where the authors directly used the phoneme label as the prediction target and form the HMM posterior with the recognized phoneme confusion matrix, and the performance reported was also better than the DNN baseline.”
Speech Enhancement with LSTM Recurrent Neural Networks and its Application to Noise-Robust ASR (2015, methodology): “Another notable finding is that stacking LSTM networks for source separation with DNNs for acoustic modeling is more promising than using LSTM networks directly for acoustic modeling: In [3], no WER gains by using LSTM acoustic models instead of DNN ones were reported on the CHiME-2 data.”
A comprehensive study of deep bidirectional LSTM RNNS for acoustic modeling in speech recognition (2016, unknown): “Index Terms— acoustic modeling, LSTM, RNN”
Maximum a posteriori Based Decoding for CTC Acoustic Models (2016, background): “Encouraged by the strong performance of DNNs, various extensions of the network architecture have been proposed, such as convolutional neural networks (CNNs) [6], sigmoid-unit-type recurrent neural networks (RNNs) [7, 8, 9, 10], and long short-term memory neural networks [11, 12, 13, 14, 15].”