Ensemble deep learning in speech signal tasks: a review
Abstract Machine learning methods are extensively used for processing and analysing
speech signals by virtue of their performance gains over multiple domains. Deep learning …
speech signals by virtue of their performance gains over multiple domains. Deep learning …
[图书][B] Distant speech recognition
M Wölfel, J McDonough - 2009 - books.google.com
A complete overview of distant automatic speech recognition The performance of
conventional Automatic Speech Recognition (ASR) systems degrades dramatically as soon …
conventional Automatic Speech Recognition (ASR) systems degrades dramatically as soon …
Speech recognition and keyword spotting for low-resource languages: Babel project research at cued
Recently there has been increased interest in Automatic Speech Recognition (ASR) and
Key Word Spotting (KWS) systems for low resource languages. One of the driving forces for …
Key Word Spotting (KWS) systems for low resource languages. One of the driving forces for …
Boosted MMI for model and feature-space discriminative training
We present a modified form of the maximum mutual information (MMI) objective function
which gives improved results for discriminative training. The modification consists of …
which gives improved results for discriminative training. The modification consists of …
Audio-visual deep learning for noise robust speech recognition
J Huang, B Kingsbury - 2013 IEEE international conference on …, 2013 - ieeexplore.ieee.org
Deep belief networks (DBN) have shown impressive improvements over Gaussian mixture
models for automatic speech recognition. In this work we use DBNs for audio-visual speech …
models for automatic speech recognition. In this work we use DBNs for audio-visual speech …
Lattice-based optimization of sequence classification criteria for neural-network acoustic modeling
B Kingsbury - 2009 IEEE International Conference on Acoustics …, 2009 - ieeexplore.ieee.org
Acoustic models used in hidden Markov model/neural-network (HMM/NN) speech
recognition systems are usually trained with a frame-based cross-entropy error criterion. In …
recognition systems are usually trained with a frame-based cross-entropy error criterion. In …
An enhanced stacked LSTM method with no random initialization for malware threat hunting in safety and time-critical systems
AN Jahromi, S Hashemi… - … on Emerging Topics …, 2020 - ieeexplore.ieee.org
Malware detection is an increasingly important operational focus in cyber security,
particularly, given the fast pace of such threats (eg, new malware variants introduced every …
particularly, given the fast pace of such threats (eg, new malware variants introduced every …
[PDF][PDF] Scalable Minimum Bayes Risk Training of Deep Neural Network Acoustic Models Using Distributed Hessian-free Optimization.
Training neural network acoustic models with sequencediscriminative criteria, such as state-
level minimum Bayes risk (sMBR), been shown to produce large improvements in …
level minimum Bayes risk (sMBR), been shown to produce large improvements in …
[PDF][PDF] Hidden conditional random fields for phone classification.
In this paper, we show the novel application of hidden conditional random fields (HCRFs)–
conditional random fields with hidden state sequences–for modeling speech. Hidden state …
conditional random fields with hidden state sequences–for modeling speech. Hidden state …
Maximum F1-score discriminative training criterion for automatic mispronunciation detection
We carry out an in-depth investigation on a newly proposed Maximum F1-score Criterion
(MFC) discriminative training objective function for Goodness of Pronunciation (GOP) based …
(MFC) discriminative training objective function for Goodness of Pronunciation (GOP) based …