# llu.bib

@article{lu2013,
author = {Lu, Liang and Chin, KK and Ghoshal, Arnab and Renals, Steve},
doi = {10.1109/TASL.2013.2248718},
title = {Joint Uncertainty Decoding for Noise Robust Subspace {Gaussian} Mixture Models},
journal = {IEEE Transactions on Audio, Speech and Language Processing},
number = {9},
abstract = {Joint uncertainty decoding (JUD) is a model-based noise compensation technique for conventional Gaussian Mixture Model (GMM) based speech recognition systems. Unlike vector Taylor series (VTS) compensation which operates on the individual Gaussian components in an acoustic model, JUD clusters the Gaussian components into a smaller number of classes, sharing the compensation parameters for the set of Gaussians in a given class. This significantly reduces the computational cost. In this paper, we investigate noise compensation for subspace Gaussian mixture model (SGMM) based speech recognition systems using JUD. The total number of Gaussian components in an SGMM is typically very large. Therefore direct compensation of the individual Gaussian components, as performed by VTS, is computationally expensive. In this paper we show that JUD-based noise compensation can be successfully applied to SGMMs in a computationally efficient way. We evaluate the JUD/SGMM technique on the standard Aurora 4 corpus. Our experimental results indicate that the JUD/SGMM system results in lower word error rates compared with a conventional GMM system with either VTS-based or JUD-based noise compensation.},
volume = {21},
year = {2013},
pages = {1791--1804}
}

@inproceedings{6423522,
author = {Yang, Chen-Yu and Brown, G. and Lu, Liang and Yamagishi, J. and King, S.},
doi = {10.1109/ISCSLP.2012.6423522},
title = {Noise-robust whispered speech recognition using a non-audible-murmur microphone with VTS compensation},
abstract = {In this paper, we introduce a newly-created corpus of whispered speech simultaneously recorded via a close-talking microphone and a non-audible murmur (NAM) microphone in both clean and noisy conditions. To benchmark the corpus, which has been freely released recently, experiments on automatic recognition of continuous whispered speech were conducted. When training and test conditions are matched, the NAM microphone is found to be more robust against background noise than the close-talking microphone. In mismatched conditions (noisy data, models trained on clean speech), we found that Vector Taylor Series (VTS) compensation is particularly effective for the NAM signal.},
year = {2012},
booktitle = {Chinese Spoken Language Processing (ISCSLP), 2012 8th International Symposium on},
pages = {220-223}
}

@inproceedings{llu2012map,
author = {Lu, L. and Ghoshal, A. and Renals, S.},
doi = {10.1109/ICASSP.2012.6289012},
title = {{Maximum a posteriori adaptation of subspace Gaussian mixture models for cross-lingual speech recognition}},
booktitle = {Proc. ICASSP},
abstract = {This paper concerns cross-lingual acoustic modeling in the case when there are limited target language resources. We build on an approach in which a subspace Gaussian mixture model (SGMM) is adapted to the target language by reusing the globally shared parameters estimated from out-of-language training data. In current cross-lingual systems, these parameters are fixed when training the target system, which can give rise to a mismatch between the source and target systems. We investigate a maximum a posteriori (MAP) adaptation approach to alleviate the potential mismatch. In particular, we focus on the adaptation of phonetic subspace parameters using a matrix variate Gaussian prior distribution. Experiments on the GlobalPhone corpus using the MAP adaptation approach results in word error rate reductions, compared with the cross-lingual baseline systems and systems updated using maximum likelihood, for training conditions with 1 hour and 5 hours of target language data.},
year = {2012},
keywords = {Subspace Gaussian Mixture Model, Maximum a Posteriori Adaptation, Cross-lingual Speech Recognition},
pages = {4877--4880}
}

@article{lu_spl_2011,
author = {Lu, L. and Ghoshal, A. and Renals, S.},
title = {Regularized Subspace Gausian Mixture Models for Speech Recognition},
journal = {IEEE Signal Processing Letters},
number = {7},
abstract = {Subspace Gaussian mixture models (SGMMs) provide a compact representation of the Gaussian parameters in an acoustic model, but may still suffer from over-fitting with insufficient training data. In this letter, the SGMM state parameters are estimated using a penalized maximum-likelihood objective, based on $\ell_1$ and $\ell_2$ regularization, as well as their combination, referred to as the elastic net, for robust model estimation. Experiments on the 5000-word Wall Street Journal transcription task show word error rate reduction and improved model robustness with regularization.},
volume = {18},
year = {2011},
pages = {419--422},
categories = {Acoustic Modelling, Regularization, Sparsity, Subspace Gaussian Mixture Model}
}

@inproceedings{lu2012jud,
author = {Lu, L. and Ghoshal, A. and Renals, S.},
title = {{Joint uncertainty decoding with unscented transform for noise robust subspace Gaussian mixture model}},
booktitle = {Proc. Sapa-Scale workshop},
year = {2012},
keywords = {noise compensation, SGMM, JUD, UT},
abstract = {Common noise compensation techniques use vector Taylor series (VTS) to approximate the mismatch function. Recent work shows that the approximation accuracy may be improved by sampling. One such sampling technique is the unscented transform (UT), which draws samples deterministically from clean speech and noise model to derive the noise corrupted speech parameters. This paper applies UT to noise compensation of the subspace Gaussian mixture model (SGMM). Since UT requires relatively smaller number of samples for accurate estimation, it has significantly lower computational cost compared to other random sampling techniques. However, the number of surface Gaussians in an SGMM is typically very large, making the direct application of UT, for compensating individual Gaussian components, computationally impractical. In this paper, we avoid the computational burden by employing UT in the framework of joint uncertainty decoding (JUD), which groups all the Gaussian components into small number of classes, sharing the compensation parameters by class. We evaluate the JUD-UT technique for an SGMM system using the Aurora 4 corpus. Experimental results indicate that UT can lead to increased accuracy compared to VTS approximation if the JUD phase factor is untuned, and to similar accuracy if the phase factor is tuned empirically}
}

@inproceedings{lu2012noise,
author = {Lu, L. and Chin, KK and Ghoshal, A. and Renals, S.},
title = {{Noise compensation for subspace Gaussian mixture models}},
booktitle = {Proc. Interspeech},
year = {2012},
keywords = {acoustic modelling, noise compensation, SGMM, JUD},
abstract = {Joint uncertainty decoding (JUD) is an effective model-based noise compensation technique for conventional Gaussian mixture model (GMM) based speech recognition systems. In this paper, we apply JUD to subspace Gaussian mixture model (SGMM) based acoustic models. The total number of Gaussians in the SGMM acoustic model is usually much larger than for conventional GMMs, which limits the application of approaches which explicitly compensate each Gaussian, such as vector Taylor series (VTS). However, by clustering the Gaussian components into a number of regression classes, JUD-based noise compensation can be successfully applied to SGMM systems. We evaluate the JUD/SGMM technique using the Aurora 4 corpus, and the experimental results indicated that it is more accurate than conventional GMM-based systems using either VTS or JUD noise compensation.}
}

@inproceedings{lu_asru_2011,
author = {Lu, L. and Ghoshal, A. and Renals, S.},
title = {Regularized Subspace {G}ausian Mixture Models for Cross-lingual Speech Recognition},
booktitle = {Proc. ASRU},
year = {2011},
abstract = {We investigate cross-lingual acoustic modelling for low resource languages using the subspace Gaussian mixture model (SGMM). We assume the presence of acoustic models trained on multiple source languages, and use the global subspace parameters from those models for improved modelling in a target language with limited amounts of transcribed speech. Experiments on the GlobalPhone corpus using Spanish, Portuguese, and Swedish as source languages and German as target language (with 1 hour and 5 hours of transcribed audio) show that multilingually trained SGMM shared parameters result in lower word error rates (WERs) than using those from a single source language. We also show that regularizing the estimation of the SGMM state vectors by penalizing their $\ell_1$-norm help to overcome numerical instabilities and lead to lower WER.},
categories = {Subspace Gaussian Mixture Model, Cross-lingual, model regularization}
}

@inproceedings{lu2013_nat,
author = {Lu, Liang and Ghoshal, Arnab and Renals, Steve},
title = {Noise adaptive training for subspace {Gaussian} mixture models},
abstract = {Noise adaptive training (NAT) is an effective approach to normalise environmental distortions when training a speech recogniser on noise-corrupted speech. This paper investigates the model-based NAT scheme using joint uncertainty decoding (JUD) for subspace Gaussian mixture models (SGMMs). A typical SGMM acoustic model has much larger number of surface Gaussian components, which makes it computationally infeasible to compensate each Gaussian explicitly. JUD tackles this problem by sharing the compensation parameters among the Gaussians and hence reduces the computational and memory demands. For noise adaptive training, JUD is reformulated into a generative model, which leads to an efficient expectation-maximisation (EM) based algorithm to update the SGMM acoustic model parameters. We evaluated the SGMMs with NAT on the Aurora 4 database, and obtained higher recognition accuracy compared to systems without adaptive training.},
year = {2013},
booktitle = {Proc. Interspeech},
categories = {adaptive training, noise robustness, joint uncertainty decoding, subspace Gaussian mixture models}
}

@inproceedings{lu2013_pronunciation,
author = {Lu, Liang and Ghoshal, Arnab and Renals, Steve},
doi = {10.1109/ASRU.2013.6707759},
title = {Acoustic Data-driven Pronunciation Lexicon for Large Vocabulary Speech Recognition},
abstract = {Speech recognition systems normally use handcrafted pronunciation lexicons designed by linguistic experts. Building and maintaining such a lexicon is expensive and time consuming. This paper concerns automatically learning a pronunciation lexicon for speech recognition. We assume the availability of a small seed lexicon and then learn the pronunciations of new words directly from speech that is transcribed at word-level. We present two implementations for refining the putative pronunciations of new words based on acoustic evidence. The first one is an expectation maximization (EM) algorithm based on weighted finite state transducers (WFSTs) and the other is its Viterbi approximation. We carried out experiments on the Switchboard corpus of conversational telephone speech. The expert lexicon has a size of more than 30,000 words, from which we randomly selected 5,000 words to form the seed lexicon. By using the proposed lexicon learning method, we have significantly improved the accuracy compared with a lexicon learned using a grapheme-to-phoneme transformation, and have obtained a word error rate that approaches that achieved using a fully handcrafted lexicon.},
year = {2013},
booktitle = {Proc. ASRU},
categories = {Lexical modelling, Probabilistic pronunciation model, Automatic speech recognition}
}

@phdthesis{lu2013_thesis,
author = {Lu, Liang},
school = {University of Edinburgh},
title = {Subspace {Gaussian} Mixture Models for Automatic Speech Recognition},
abstract = {In most of state-of-the-art speech recognition systems, Gaussian mixture models (GMMs) are used to model the density of the emitting states in the hidden Markov models (HMMs). In a conventional system, the model parameters of each GMM are estimated directly and independently given the alignment. This results a large number of model parameters to be estimated, and consequently, a large amount of training data is required to fit the model. In addition, different sources of acoustic variability that impact the accuracy of a recogniser such as pronunciation variation, accent, speaker factor and environmental noise are only weakly modelled and factorized by adaptation techniques such as maximum likelihood linear regression (MLLR), maximum a posteriori adaptation (MAP) and vocal tract length normalisation (VTLN). In this thesis, we will discuss an alternative acoustic modelling approach --- the subspace Gaussian mixture model (SGMM), which is expected to deal with these two issues better. In an SGMM, the model parameters are derived from low-dimensional model and speaker subspaces that can capture phonetic and speaker correlations. Given these subspaces, only a small number of state-dependent parameters are required to derive the corresponding GMMs. Hence, the total number of model parameters can be reduced, which allows acoustic modelling with a limited amount of training data. In addition, the SGMM-based acoustic model factorizes the phonetic and speaker factors and within this framework, other source of acoustic variability may also be explored. In this thesis, we propose a regularised model estimation for SGMMs, which avoids overtraining in case that the training data is sparse. We will also take advantage of the structure of SGMMs to explore cross-lingual acoustic modelling for low-resource speech recognition. Here, the model subspace is estimated from out-domain data and ported to the target language system. In this case, only the state-dependent parameters need to be estimated which relaxes the requirement of the amount of training data. To improve the robustness of SGMMs against environmental noise, we propose to apply the joint uncertainty decoding (JUD) technique that is shown to be efficient and effective. We will report experimental results on the Wall Street Journal (WSJ) database and GlobalPhone corpora to evaluate the regularisation and cross-lingual modelling of SGMMs. Noise compensation using JUD for SGMM acoustic models is evaluated on the Aurora 4 database.},
year = {2013},
categories = {subspace Gaussian mixture model, speech recognition, cross-lingual, noise robustness}
}

@article{lu2013cross,
author = {Lu, Liang and Ghoshal, Arnab and Renals, Steve},
doi = {10.1109/TASL.2013.2281575},
title = {{Cross-lingual subspace {Gaussian} mixture model for low-resource speech recognition}},
journal = {IEEE Transactions on Audio, Speech and Language Processing},
number = {1},
abstract = {This paper studies cross-lingual acoustic modelling in the context of subspace Gaussian mixture models (SGMMs). SGMMs factorize the acoustic model parameters into a set that is globally shared between all the states of a hidden Markov model (HMM) and another that is specific to the HMM states. We demonstrate that the SGMM global parameters are transferable between languages, particularly when the parameters are trained multilingually. As a result, acoustic models may be trained using limited amounts of transcribed audio by borrowing the SGMM global parameters from one or more source languages, and only training the state-specific parameters on the target language audio. Model regularization using $\ell_1$-norm penalty is shown to be particularly effective at avoiding overtraining and leading to lower word error rates. We investigate maximum a posteriori (MAP) adaptation of subspace parameters in order to reduce the mismatch between the SGMM global parameters of the source and target languages. In addition, monolingual and cross-lingual speaker adaptive training is used to reduce the model variance introduced by speakers. We have systematically evaluated these techniques by experiments on the GlobalPhone corpus.},
volume = {22},
year = {2014},
pages = {17--27},
categories = {acoustic modelling, subspace Gaussian mixture model, cross-lingual speech recognition, regularization, adaptation}
}

@article{lu2014a,
author = {Lu, Liang and Renals, Steve},
doi = {10.1109/LSP.2014.2313410},
title = {Probabilistic Linear Discriminant Analysis for Acoustic Modelling},
journal = {IEEE Signal Processing Letters},
number = {6},
pages = {702--706},
volume = {21},
year = {2014},
abstract = {In this letter, we propose a new acoustic modelling approach for automatic speech recognition based on probabilistic linear discriminant analysis (PLDA), which is used to model the state density function for the standard hidden Markov models (HMMs). Unlike the conventional Gaussian mixture models (GMMs) where the correlations are weakly modelled by using the diagonal covariance matrices, PLDA captures the correlations of feature vector in subspaces without vastly expanding the model. It also allows the usage of high dimensional feature input, and therefore is more flexible to make use of different type of acoustic features. We performed the preliminary experiments on the Switchboard corpus, and demonstrated the feasibility of this acoustic model.}
}

@inproceedings{llu_IS2014,
author = {Lu, Liang and Renals, Steve},
title = {Probabilistic linear discriminant analysis with bottleneck features for speech recognition},
abstract = {We have recently proposed a new acoustic model based on prob- abilistic linear discriminant analysis (PLDA) which enjoys the flexibility of using higher dimensional acoustic features, and is more capable to capture the intra-frame feature correlations. In this paper, we investigate the use of bottleneck features obtained from a deep neural network (DNN) for the PLDA-based acous- tic model. Experiments were performed on the Switchboard dataset --- a large vocabulary conversational telephone speech corpus. We observe significant word error reduction by using the bottleneck features. In addition, we have also compared the PLDA-based acoustic model to three others using Gaussian mixture models (GMMs), subspace GMMs and hybrid deep neural networks (DNNs), and PLDA can achieve comparable or slightly higher recognition accuracy from our experiments.},
year = {2014},
booktitle = {Proc. Interspeech},
categories = {speech recognition, bottleneck features, probabilistic linear discriminant analysis}
}

@inproceedings{llu_is2015b,
author = {Lu, Liang and Zhang, Xingxing and Cho, KyungHyun and Renals, Steve},
date-modified = {2015-08-19 10:22:47 +0100},
title = {A Study of the Recurrent Neural Network Encoder-Decoder for Large Vocabulary Speech Recognition},
abstract = {Deep neural networks have advanced the state-of-the-art in automatic speech recognition, when combined with hidden Markov models (HMMs). Recently there has been interest in using systems based on recurrent neural networks (RNNs) to perform sequence modelling directly, without the requirement of an HMM superstructure. In this paper, we study the RNN encoder-decoder approach for large vocabulary end-to-end speech recognition, whereby an encoder transforms a sequence of acoustic vectors into a sequence of feature representations, from which a decoder recovers a sequence of words. We investigated this approach on the Switchboard corpus using a training set of around 300 hours of transcribed audio data. Without the use of an explicit language model or pronunciation lexicon, we achieved promising recognition accuracy, demonstrating that this approach warrants further investigation.},
year = {2015},
booktitle = {Proc. Interspeech},
categories = {end-to-end speech recognition, deep neural networks, recurrent neural networks, encoder-decoder},
}

@inproceedings{llu_is2015a,
author = {Lu, Liang and Renals, Steve},
date-modified = {2015-08-19 10:13:33 +0100},
title = {Feature-space Speaker Adaptation for Probabilistic Linear Discriminant Analysis Acoustic Models},
abstract = {Probabilistic linear discriminant analysis (PLDA) acoustic models extend Gaussian mixture models by factorizing the acoustic variability using state-dependent and observation-dependent variables. This enables the use of higher dimensional acoustic features, and the capture of intra-frame feature correlations. In this paper, we investigate the estimation of speaker adaptive feature-space (constrained) maximum likelihood linear regression transforms from PLDA-based acoustic models. This feature-space speaker transformation estimation approach is potentially very useful due to the ability of PLDA acoustic models to use different types of acoustic features, for example applying these transforms to deep neural network (DNN) acoustic models for cross adaptation. We evaluated the approach on the Switchboard corpus, and observe significant word error reduction by using both the mel-frequency cepstral coefficients and DNN bottleneck features.},
year = {2015},
booktitle = {Proc. Interspeech},
categories = {speech recognition, probabilistic linear discriminant analysis, speaker adaptation, fMLLR, PLDA},
}

@inproceedings{llu_icassp15,
author = {Lu, Liang and Renals, Steve},
date-modified = {2015-08-19 10:16:24 +0100},
title = {Multi-frame factorisation for long-span acoustic modelling},
abstract = {Acoustic models based on Gaussian mixture models (GMMs) typically use short span acoustic feature inputs. This does not capture long-term temporal information from speech owing to the conditional independence assumption of hidden Markov models. In this paper, we present an implicit approach that approximates the joint distribution of long span features by product of factorized models, in contrast to deep neural networks (DNNs) that model feature correlations directly. The approach is applicable to a broad range of acoustic models. We present experiments using GMM and probabilistic linear discriminant analysis (PLDA) based models on Switchboard, observing consistent word error rate reductions.},
year = {2015},