Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707737
H. Konno, Hideo Kanemitsu, N. Takahashi, Mineichi Kudo
The characteristics of whispered speech are not well known. The most remarkable difference from ordinal speech is the pitch (the height of speech), since whispered speech has no fundamental frequency. In this study, we have tried to reveal the mechanism of producing pitch in whispered speech through an experiment in which a male and a female subjects uttered Japanese whispered vowels in a way so as to tune their pitch to the guidance tone with different five to nine frequencies. We applied multivariate analysis such as the principal component analysis to the data in order to make clear which part of frequency contributes much to the change of pitch. We have succeeded in endorsing the previous observations, i.e. shift of formants is dominant, with more detailed numerical evidence. In addition, we obtained some implications to approach the pitch mechanism of whispered speech. The main result obtained is that two or three formants of less than 5 kHz are shifted upward and the energy is increased in high frequency region over 5 kHz.
{"title":"Acoustic characteristics related to the perceptual pitch in whispered vowels","authors":"H. Konno, Hideo Kanemitsu, N. Takahashi, Mineichi Kudo","doi":"10.1109/ASRU.2013.6707737","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707737","url":null,"abstract":"The characteristics of whispered speech are not well known. The most remarkable difference from ordinal speech is the pitch (the height of speech), since whispered speech has no fundamental frequency. In this study, we have tried to reveal the mechanism of producing pitch in whispered speech through an experiment in which a male and a female subjects uttered Japanese whispered vowels in a way so as to tune their pitch to the guidance tone with different five to nine frequencies. We applied multivariate analysis such as the principal component analysis to the data in order to make clear which part of frequency contributes much to the change of pitch. We have succeeded in endorsing the previous observations, i.e. shift of formants is dominant, with more detailed numerical evidence. In addition, we obtained some implications to approach the pitch mechanism of whispered speech. The main result obtained is that two or three formants of less than 5 kHz are shifted upward and the energy is increased in high frequency region over 5 kHz.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121139059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707724
Antti Hurmalainen, T. Virtanen
Non-negative spectral factorisation with long temporal context has been successfully used for noise robust recognition of speech in multi-source environments. Sparse classification from activations of speech atoms can be employed instead of conventional GMMs to determine speech state likelihoods. For accurate classification, correct linguistic state labels must be assigned to speech atoms. We propose using non-negative matrix deconvolution for learning the labels with algorithms closely matching a framework that separates speech from additive noises. Experiments on the 1st CHiME Challenge corpus show improvement in recognition accuracy over labels acquired from original atom sources or previously used least squares regression. The new approach also circumvents numerical issues encountered in previous learning methods, and opens up possibilities for new speech basis generation algorithms.
{"title":"Learning state labels for sparse classification of speech with matrix deconvolution","authors":"Antti Hurmalainen, T. Virtanen","doi":"10.1109/ASRU.2013.6707724","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707724","url":null,"abstract":"Non-negative spectral factorisation with long temporal context has been successfully used for noise robust recognition of speech in multi-source environments. Sparse classification from activations of speech atoms can be employed instead of conventional GMMs to determine speech state likelihoods. For accurate classification, correct linguistic state labels must be assigned to speech atoms. We propose using non-negative matrix deconvolution for learning the labels with algorithms closely matching a framework that separates speech from additive noises. Experiments on the 1st CHiME Challenge corpus show improvement in recognition accuracy over labels acquired from original atom sources or previously used least squares regression. The new approach also circumvents numerical issues encountered in previous learning methods, and opens up possibilities for new speech basis generation algorithms.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129252356","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707722
N. Prasad, S. Umesh
Cepstral Mean and Variance Normalization (CMVN) is a computationally efficient normalization technique for noise robust speech recognition. The performance of CMVN is known to degrade for short utterances, due to insufficient data for parameter estimation and loss of discriminable information as all utterances are forced to have zero mean and unit variance. In this work, we propose to use posterior estimates of mean and variance in CMVN, instead of the maximum likelihood estimates. This Bayesian approach, in addition to providing a robust estimate of parameters, is also shown to preserve discriminable information without increase in computational cost, making it particularly relevant for Interactive Voice Response (IVR)-based applications. The relative WER reduction of this approach w.r.t. Cepstral Mean Normalization, CMVN and Histogram Equalization are (i) 40.1%, 27% and 4.3% with the Aurora2 database for all utterances, (ii) 25.7%, 38.6% and 30.4% with the Aurora2 database for short utterances, and (iii) 18.7%, 12.6% and 2.5% with the Aurora4 database.
{"title":"Improved cepstral mean and variance normalization using Bayesian framework","authors":"N. Prasad, S. Umesh","doi":"10.1109/ASRU.2013.6707722","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707722","url":null,"abstract":"Cepstral Mean and Variance Normalization (CMVN) is a computationally efficient normalization technique for noise robust speech recognition. The performance of CMVN is known to degrade for short utterances, due to insufficient data for parameter estimation and loss of discriminable information as all utterances are forced to have zero mean and unit variance. In this work, we propose to use posterior estimates of mean and variance in CMVN, instead of the maximum likelihood estimates. This Bayesian approach, in addition to providing a robust estimate of parameters, is also shown to preserve discriminable information without increase in computational cost, making it particularly relevant for Interactive Voice Response (IVR)-based applications. The relative WER reduction of this approach w.r.t. Cepstral Mean Normalization, CMVN and Histogram Equalization are (i) 40.1%, 27% and 4.3% with the Aurora2 database for all utterances, (ii) 25.7%, 38.6% and 30.4% with the Aurora2 database for short utterances, and (iii) 18.7%, 12.6% and 2.5% with the Aurora4 database.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121416370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707735
A. Fuchs, J. A. Morales-Cordovilla, Martin Hagmüller
The electro-larynx device (EL) offers the possibility to re-obtain speech when the larynx is removed after a total laryngectomy. Speech produced with an EL suffers from inadequate speech sound quality, therefore there is a strong need to enhance EL speech. When disordered speech is applied to Automatic Speech Recognition (ASR) systems, the performance will significantly decrease. ASR systems are increasingly part of daily life and therefore, the word accuracy rate of disordered speech should be reasonably high in order to be able to make ASR technologies accessible for patients suffering from speech disorders. Moreover, ASR is a method to get an objective rating for the intelligibility of disordered speech. In this paper we apply disordered speech, namely speech produced by an EL, on an ASR system which was designed for normal, healthy speech and evaluate its performance with different types of adaptation. Furthermore, we show that two approaches to reduce the directly radiated EL (DREL) noise from the device itself are able to increase the word accuracy rate compared to the unprocessed EL speech.
{"title":"ASR for electro-laryngeal speech","authors":"A. Fuchs, J. A. Morales-Cordovilla, Martin Hagmüller","doi":"10.1109/ASRU.2013.6707735","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707735","url":null,"abstract":"The electro-larynx device (EL) offers the possibility to re-obtain speech when the larynx is removed after a total laryngectomy. Speech produced with an EL suffers from inadequate speech sound quality, therefore there is a strong need to enhance EL speech. When disordered speech is applied to Automatic Speech Recognition (ASR) systems, the performance will significantly decrease. ASR systems are increasingly part of daily life and therefore, the word accuracy rate of disordered speech should be reasonably high in order to be able to make ASR technologies accessible for patients suffering from speech disorders. Moreover, ASR is a method to get an objective rating for the intelligibility of disordered speech. In this paper we apply disordered speech, namely speech produced by an EL, on an ASR system which was designed for normal, healthy speech and evaluate its performance with different types of adaptation. Furthermore, we show that two approaches to reduce the directly radiated EL (DREL) noise from the device itself are able to increase the word accuracy rate compared to the unprocessed EL speech.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127877663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707721
Rongfeng Su, Xunying Liu, Lan Wang
An important task for speech recognition systems is to handle the mismatch against a target environment introduced by acoustic factors such as variable ambient noise. To address this issue, it is possible to explicitly approximate the continuous trajectory of optimal, well matched model parameters against the varying noise using, for example, using generalized variable parameter HMMs (GVP-HMM). In order to improve the generalization and computational efficiency of conventional GVP-HMMs, this paper investigates a novel model complexity control method for GVP-HMMs. The optimal polynomial degrees of Gaussian mean, variance and model space linear transform trajectories are automatically determined at local level. Significant error rate reductions of 20% and 28% relative were obtained over the multi-style training baseline systems on Aurora 2 and a medium vocabulary Mandarin Chinese speech recognition task respectively. Consistent performance improvements and model size compression of 57% relative were also obtained over the baseline GVP-HMM systems using a uniformly assigned polynomial degree.
{"title":"Automatic model complexity control for generalized variable parameter HMMs","authors":"Rongfeng Su, Xunying Liu, Lan Wang","doi":"10.1109/ASRU.2013.6707721","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707721","url":null,"abstract":"An important task for speech recognition systems is to handle the mismatch against a target environment introduced by acoustic factors such as variable ambient noise. To address this issue, it is possible to explicitly approximate the continuous trajectory of optimal, well matched model parameters against the varying noise using, for example, using generalized variable parameter HMMs (GVP-HMM). In order to improve the generalization and computational efficiency of conventional GVP-HMMs, this paper investigates a novel model complexity control method for GVP-HMMs. The optimal polynomial degrees of Gaussian mean, variance and model space linear transform trajectories are automatically determined at local level. Significant error rate reductions of 20% and 28% relative were obtained over the multi-style training baseline systems on Aurora 2 and a medium vocabulary Mandarin Chinese speech recognition task respectively. Consistent performance improvements and model size compression of 57% relative were also obtained over the baseline GVP-HMM systems using a uniformly assigned polynomial degree.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117024410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707715
Takuya Hiraoka, Yuki Yamauchi, Graham Neubig, S. Sakti, T. Toda, Satoshi Nakamura
In this research, we propose a probabilistic dialogue modeling method for persuasive dialogue systems that interact with the user based on a specific goal, and lead the user to take actions that the system intends from candidate actions satisfying the user's needs. As a baseline system, we develop a dialogue model assuming the user makes decisions based on preference. Then we improve the model by introducing methods to guide the user from topic to topic. We evaluate the system knowledge and dialogue manager in a task that tests the system's persuasive power, and find that the proposed method is effective in this respect.
{"title":"Dialogue management for leading the conversation in persuasive dialogue systems","authors":"Takuya Hiraoka, Yuki Yamauchi, Graham Neubig, S. Sakti, T. Toda, Satoshi Nakamura","doi":"10.1109/ASRU.2013.6707715","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707715","url":null,"abstract":"In this research, we propose a probabilistic dialogue modeling method for persuasive dialogue systems that interact with the user based on a specific goal, and lead the user to take actions that the system intends from candidate actions satisfying the user's needs. As a baseline system, we develop a dialogue model assuming the user makes decisions based on preference. Then we improve the model by introducing methods to guide the user from topic to topic. We evaluate the system knowledge and dialogue manager in a task that tests the system's persuasive power, and find that the proposed method is effective in this respect.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123582167","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707705
G. Saon, H. Soltau, D. Nahamoo, M. Picheny
We propose to adapt deep neural network (DNN) acoustic models to a target speaker by supplying speaker identity vectors (i-vectors) as input features to the network in parallel with the regular acoustic features for ASR. For both training and test, the i-vector for a given speaker is concatenated to every frame belonging to that speaker and changes across different speakers. Experimental results on a Switchboard 300 hours corpus show that DNNs trained on speaker independent features and i-vectors achieve a 10% relative improvement in word error rate (WER) over networks trained on speaker independent features only. These networks are comparable in performance to DNNs trained on speaker-adapted features (with VTLN and FMLLR) with the advantage that only one decoding pass is needed. Furthermore, networks trained on speaker-adapted features and i-vectors achieve a 5-6% relative improvement in WER after hessian-free sequence training over networks trained on speaker-adapted features only.
{"title":"Speaker adaptation of neural network acoustic models using i-vectors","authors":"G. Saon, H. Soltau, D. Nahamoo, M. Picheny","doi":"10.1109/ASRU.2013.6707705","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707705","url":null,"abstract":"We propose to adapt deep neural network (DNN) acoustic models to a target speaker by supplying speaker identity vectors (i-vectors) as input features to the network in parallel with the regular acoustic features for ASR. For both training and test, the i-vector for a given speaker is concatenated to every frame belonging to that speaker and changes across different speakers. Experimental results on a Switchboard 300 hours corpus show that DNNs trained on speaker independent features and i-vectors achieve a 10% relative improvement in word error rate (WER) over networks trained on speaker independent features only. These networks are comparable in performance to DNNs trained on speaker-adapted features (with VTLN and FMLLR) with the advantage that only one decoding pass is needed. Furthermore, networks trained on speaker-adapted features and i-vectors achieve a 5-6% relative improvement in WER after hessian-free sequence training over networks trained on speaker-adapted features only.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121036310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707745
Meng Cai, Yongzhe Shi, Jia Liu
A recently introduced type of neural network called maxout has worked well in many domains. In this paper, we propose to apply maxout for acoustic models in speech recognition. The maxout neuron picks the maximum value within a group of linear pieces as its activation. This nonlinearity is a generalization to the rectified nonlinearity and has the ability to approximate any form of activation functions. We apply maxout networks to the Switchboard phone-call transcription task and evaluate the performances under both a 24-hour low-resource condition and a 300-hour core condition. Experimental results demonstrate that maxout networks converge faster, generalize better and are easier to optimize than rectified linear networks and sigmoid networks. Furthermore, experiments show that maxout networks reduce underfitting and are able to achieve good results without dropout training. Under both conditions, maxout networks yield relative improvements of 1.1-5.1% over rectified linear networks and 2.6-14.5% over sigmoid networks on benchmark test sets.
{"title":"Deep maxout neural networks for speech recognition","authors":"Meng Cai, Yongzhe Shi, Jia Liu","doi":"10.1109/ASRU.2013.6707745","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707745","url":null,"abstract":"A recently introduced type of neural network called maxout has worked well in many domains. In this paper, we propose to apply maxout for acoustic models in speech recognition. The maxout neuron picks the maximum value within a group of linear pieces as its activation. This nonlinearity is a generalization to the rectified nonlinearity and has the ability to approximate any form of activation functions. We apply maxout networks to the Switchboard phone-call transcription task and evaluate the performances under both a 24-hour low-resource condition and a 300-hour core condition. Experimental results demonstrate that maxout networks converge faster, generalize better and are easier to optimize than rectified linear networks and sigmoid networks. Furthermore, experiments show that maxout networks reduce underfitting and are able to achieve good results without dropout training. Under both conditions, maxout networks yield relative improvements of 1.1-5.1% over rectified linear networks and 2.6-14.5% over sigmoid networks on benchmark test sets.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127010455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707733
Han-Ping Shen, N. Minematsu, T. Makino, S. Weinberger, T. Pongkittiphan, Chung-Hsien Wu
English is the only language available for global communication. Due to the influence of speakers' mother tongue, however, those from different regions inevitably have different accents in their pronunciation of English. The ultimate goal of our project is creating a global pronunciation map of World Englishes on an individual basis, for speakers to use to locate similar English pronunciations. If the speaker is a learner, he can also know how his pronunciation compares to other varieties. Creating the map mathematically requires a matrix of pronunciation distances among all the speakers considered. This paper investigates invariant pronunciation structure analysis and Support Vector Regression (SVR) to predict the inter-speaker pronunciation distances. In experiments, the Speech Accent Archive (SAA), which contains speech data of worldwide accented English, is used as training and testing samples. IPA narrow transcriptions in the archive are used to prepare reference pronunciation distances, which are then predicted based on structural analysis and SVR, not with IPA transcriptions. Correlation between the reference distances and the predicted distances is calculated. Experimental results show very promising results and our proposed method outperforms by far a baseline system developed using an HMM-based phoneme recognizer.
{"title":"Automatic pronunciation clustering using a World English archive and pronunciation structure analysis","authors":"Han-Ping Shen, N. Minematsu, T. Makino, S. Weinberger, T. Pongkittiphan, Chung-Hsien Wu","doi":"10.1109/ASRU.2013.6707733","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707733","url":null,"abstract":"English is the only language available for global communication. Due to the influence of speakers' mother tongue, however, those from different regions inevitably have different accents in their pronunciation of English. The ultimate goal of our project is creating a global pronunciation map of World Englishes on an individual basis, for speakers to use to locate similar English pronunciations. If the speaker is a learner, he can also know how his pronunciation compares to other varieties. Creating the map mathematically requires a matrix of pronunciation distances among all the speakers considered. This paper investigates invariant pronunciation structure analysis and Support Vector Regression (SVR) to predict the inter-speaker pronunciation distances. In experiments, the Speech Accent Archive (SAA), which contains speech data of worldwide accented English, is used as training and testing samples. IPA narrow transcriptions in the archive are used to prepare reference pronunciation distances, which are then predicted based on structural analysis and SVR, not with IPA transcriptions. Correlation between the reference distances and the predicted distances is calculated. Experimental results show very promising results and our proposed method outperforms by far a baseline system developed using an HMM-based phoneme recognizer.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116496012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707744
P. Swietojanski, Arnab Ghoshal, S. Renals
We investigate the application of deep neural network (DNN)-hidden Markov model (HMM) hybrid acoustic models for far-field speech recognition of meetings recorded using microphone arrays. We show that the hybrid models achieve significantly better accuracy than conventional systems based on Gaussian mixture models (GMMs). We observe up to 8% absolute word error rate (WER) reduction from a discriminatively trained GMM baseline when using a single distant microphone, and between 4-6% absolute WER reduction when using beamforming on various combinations of array channels. By training the networks on audio from multiple channels, we find the networks can recover significant part of accuracy difference between the single distant microphone and beamformed configurations. Finally, we show that the accuracy of a network recognising speech from a single distant microphone can approach that of a multi-microphone setup by training with data from other microphones.
{"title":"Hybrid acoustic models for distant and multichannel large vocabulary speech recognition","authors":"P. Swietojanski, Arnab Ghoshal, S. Renals","doi":"10.1109/ASRU.2013.6707744","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707744","url":null,"abstract":"We investigate the application of deep neural network (DNN)-hidden Markov model (HMM) hybrid acoustic models for far-field speech recognition of meetings recorded using microphone arrays. We show that the hybrid models achieve significantly better accuracy than conventional systems based on Gaussian mixture models (GMMs). We observe up to 8% absolute word error rate (WER) reduction from a discriminatively trained GMM baseline when using a single distant microphone, and between 4-6% absolute WER reduction when using beamforming on various combinations of array channels. By training the networks on audio from multiple channels, we find the networks can recover significant part of accuracy difference between the single distant microphone and beamformed configurations. Finally, we show that the accuracy of a network recognising speech from a single distant microphone can approach that of a multi-microphone setup by training with data from other microphones.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133626109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：