Pub Date : 2016-12-01DOI: 10.1109/SLT.2016.7846316
Richard Dufour, Mohamed Morchid, Titouan Parcollet
Automatically translating textual documents from one language to another inevitably results in translation errors. In addition to language specificities, this automatic translation appears more difficult in the context of spoken dialogues since, for example, the language register is far from “clean speech”. Speech analytics suffer from these translation errors. To tackle this difficulty, a solution consists in mapping translations into a space of hidden topics. In the classical topic-based representation obtained from a Latent Dirichlet Allocation (LDA), distribution of words into each topic is estimated automatically. Nonetheless, the targeted classes are ignored in the particular context of a classification task. In the DSTC5 main task, this targeted class information is crucial, the main objective being to track dialog states for sub-dialog segments. For this challenge, we propose to apply an original topic-based representation for each sub-dialogue based not only on the sub-dialogue content itself (words), but also on the dialogue state related to the sub-dialogue. This original representation is based on the Author-Topic (AT) model, previously successfully applied on a different classification task. Promising results confirmed the interest of such a method, the AT model reaching performance slightly better in terms of F-measure than baseline ones given by the task's organizers.
{"title":"Tracking dialog states using an Author-Topic based representation","authors":"Richard Dufour, Mohamed Morchid, Titouan Parcollet","doi":"10.1109/SLT.2016.7846316","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846316","url":null,"abstract":"Automatically translating textual documents from one language to another inevitably results in translation errors. In addition to language specificities, this automatic translation appears more difficult in the context of spoken dialogues since, for example, the language register is far from “clean speech”. Speech analytics suffer from these translation errors. To tackle this difficulty, a solution consists in mapping translations into a space of hidden topics. In the classical topic-based representation obtained from a Latent Dirichlet Allocation (LDA), distribution of words into each topic is estimated automatically. Nonetheless, the targeted classes are ignored in the particular context of a classification task. In the DSTC5 main task, this targeted class information is crucial, the main objective being to track dialog states for sub-dialog segments. For this challenge, we propose to apply an original topic-based representation for each sub-dialogue based not only on the sub-dialogue content itself (words), but also on the dialogue state related to the sub-dialogue. This original representation is based on the Author-Topic (AT) model, previously successfully applied on a different classification task. Promising results confirmed the interest of such a method, the AT model reaching performance slightly better in terms of F-measure than baseline ones given by the task's organizers.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128446816","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 : 2016-12-01DOI: 10.1109/SLT.2016.7846300
Ondrej Klejch, P. Bell, S. Renals
In this paper we investigate the punctuated transcription of multi-genre broadcast media. We examine four systems, three of which are based on lexical features, the fourth of which uses acoustic features by integrating punctuation into the speech recognition acoustic models. We also explore the combination of these component systems using voting and log-linear interpolation. We performed experiments on the English language MGB Challenge data, which comprises about 1,600h of BBC television recordings. Our results indicate that a lexical system, based on a neural machine translation approach is significantly better than other systems achieving an F-Measure of 62.6% on reference text, with a relative degradation of 19% on ASR output. Our analysis of the results in terms of specific punctuation indicated that using longer context improves the prediction of question marks and acoustic information improves prediction of exclamation marks. Finally, we show that even though the systems are complementary, their straightforward combination does not yield better F-measures than a single system using neural machine translation.
{"title":"Punctuated transcription of multi-genre broadcasts using acoustic and lexical approaches","authors":"Ondrej Klejch, P. Bell, S. Renals","doi":"10.1109/SLT.2016.7846300","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846300","url":null,"abstract":"In this paper we investigate the punctuated transcription of multi-genre broadcast media. We examine four systems, three of which are based on lexical features, the fourth of which uses acoustic features by integrating punctuation into the speech recognition acoustic models. We also explore the combination of these component systems using voting and log-linear interpolation. We performed experiments on the English language MGB Challenge data, which comprises about 1,600h of BBC television recordings. Our results indicate that a lexical system, based on a neural machine translation approach is significantly better than other systems achieving an F-Measure of 62.6% on reference text, with a relative degradation of 19% on ASR output. Our analysis of the results in terms of specific punctuation indicated that using longer context improves the prediction of question marks and acoustic information improves prediction of exclamation marks. Finally, we show that even though the systems are complementary, their straightforward combination does not yield better F-measures than a single system using neural machine translation.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129148739","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 : 2016-12-01DOI: 10.1109/SLT.2016.7846318
Hongjie Shi, Takashi Ushio, M. Endo, K. Yamagami, Noriaki Horii
The fifth Dialog State Tracking Challenge (DSTC5) introduces a new cross-language dialog state tracking scenario, where the participants are asked to build their trackers based on the English training corpus, while evaluating them with the unlabeled Chinese corpus. Although the computer-generated translations for both English and Chinese corpus are provided in the dataset, these translations contain errors and careless use of them can easily hurt the performance of the built trackers. To address this problem, we propose a multichannel Convolutional Neural Networks (CNN) architecture, in which we treat English and Chinese language as different input channels of one single CNN model. In the evaluation of DSTC5, we found that such multichannel architecture can effectively improve the robustness against translation errors. Additionally, our method for DSTC5 is purely machine learning based and requires no prior knowledge about the target language. We consider this a desirable property for building a tracker in the cross-language context, as not every developer will be familiar with both languages.
{"title":"A multichannel convolutional neural network for cross-language dialog state tracking","authors":"Hongjie Shi, Takashi Ushio, M. Endo, K. Yamagami, Noriaki Horii","doi":"10.1109/SLT.2016.7846318","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846318","url":null,"abstract":"The fifth Dialog State Tracking Challenge (DSTC5) introduces a new cross-language dialog state tracking scenario, where the participants are asked to build their trackers based on the English training corpus, while evaluating them with the unlabeled Chinese corpus. Although the computer-generated translations for both English and Chinese corpus are provided in the dataset, these translations contain errors and careless use of them can easily hurt the performance of the built trackers. To address this problem, we propose a multichannel Convolutional Neural Networks (CNN) architecture, in which we treat English and Chinese language as different input channels of one single CNN model. In the evaluation of DSTC5, we found that such multichannel architecture can effectively improve the robustness against translation errors. Additionally, our method for DSTC5 is purely machine learning based and requires no prior knowledge about the target language. We consider this a desirable property for building a tracker in the cross-language context, as not every developer will be familiar with both languages.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126160523","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 : 2016-12-01DOI: 10.1109/SLT.2016.7846271
Andrea Schnall, M. Heckmann
Prosodic cues are an important part of human communication. One of these cues is the word prominence which is used to e.g. highlight important information. Since individual speakers use different ways of expressing prominence, it is not easily extracted and incorporated in a dialog system. As a consequence, up to date prominence only plays a marginal role in human-machine communication. In this paper we compare DNNs and SVMs trained speaker independently with the results of classification with SVM using a speaker adaptation method we recently developed. This adaptation method is based on the radial basis function of the SVM with a Gaussian regularization, which is derived from fMLLR. With this adaptation, we can notably reduce the problem of speaker variations. We present detailed evaluations of the methods and discuss advantages and shortcomings of the proposed approaches for word prominence detection.
{"title":"Comparing speaker independent and speaker adapted classification for word prominence detection","authors":"Andrea Schnall, M. Heckmann","doi":"10.1109/SLT.2016.7846271","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846271","url":null,"abstract":"Prosodic cues are an important part of human communication. One of these cues is the word prominence which is used to e.g. highlight important information. Since individual speakers use different ways of expressing prominence, it is not easily extracted and incorporated in a dialog system. As a consequence, up to date prominence only plays a marginal role in human-machine communication. In this paper we compare DNNs and SVMs trained speaker independently with the results of classification with SVM using a speaker adaptation method we recently developed. This adaptation method is based on the radial basis function of the SVM with a Gaussian regularization, which is derived from fMLLR. With this adaptation, we can notably reduce the problem of speaker variations. We present detailed evaluations of the methods and discuss advantages and shortcomings of the proposed approaches for word prominence detection.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"73 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113971843","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 : 2016-11-08DOI: 10.1109/SLT.2016.7846310
Shane Settle, Karen Livescu
Acoustic word embeddings — fixed-dimensional vector representations of variable-length spoken word segments — have begun to be considered for tasks such as speech recognition and query-by-example search. Such embeddings can be learned discriminatively so that they are similar for speech segments corresponding to the same word, while being dissimilar for segments corresponding to different words. Recent work has found that acoustic word embeddings can outperform dynamic time warping on query-by-example search and related word discrimination tasks. However, the space of embedding models and training approaches is still relatively unexplored. In this paper we present new discriminative embedding models based on recurrent neural networks (RNNs). We consider training losses that have been successful in prior work, in particular a cross entropy loss for word classification and a contrastive loss that explicitly aims to separate same-word and different-word pairs in a “Siamese network” training setting. We find that both classifier-based and Siamese RNN embeddings improve over previously reported results on a word discrimination task, with Siamese RNNs outperforming classification models. In addition, we present analyses of the learned embeddings and the effects of variables such as dimensionality and network structure.
{"title":"Discriminative acoustic word embeddings: Tecurrent neural network-based approaches","authors":"Shane Settle, Karen Livescu","doi":"10.1109/SLT.2016.7846310","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846310","url":null,"abstract":"Acoustic word embeddings — fixed-dimensional vector representations of variable-length spoken word segments — have begun to be considered for tasks such as speech recognition and query-by-example search. Such embeddings can be learned discriminatively so that they are similar for speech segments corresponding to the same word, while being dissimilar for segments corresponding to different words. Recent work has found that acoustic word embeddings can outperform dynamic time warping on query-by-example search and related word discrimination tasks. However, the space of embedding models and training approaches is still relatively unexplored. In this paper we present new discriminative embedding models based on recurrent neural networks (RNNs). We consider training losses that have been successful in prior work, in particular a cross entropy loss for word classification and a contrastive loss that explicitly aims to separate same-word and different-word pairs in a “Siamese network” training setting. We find that both classifier-based and Siamese RNN embeddings improve over previously reported results on a word discrimination task, with Siamese RNNs outperforming classification models. In addition, we present analyses of the learned embeddings and the effects of variables such as dimensionality and network structure.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114447052","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 : 2016-10-21DOI: 10.1109/SLT.2016.7846309
Hao Tang, Weiran Wang, Kevin Gimpel, Karen Livescu
Recent work on discriminative segmental models has shown that they can achieve competitive speech recognition performance, using features based on deep neural frame classifiers. However, segmental models can be more challenging to train than standard frame-based approaches. While some segmental models have been successfully trained end to end, there is a lack of understanding of their training under different settings and with different losses.
{"title":"End-to-end training approaches for discriminative segmental models","authors":"Hao Tang, Weiran Wang, Kevin Gimpel, Karen Livescu","doi":"10.1109/SLT.2016.7846309","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846309","url":null,"abstract":"Recent work on discriminative segmental models has shown that they can achieve competitive speech recognition performance, using features based on deep neural frame classifiers. However, segmental models can be more challenging to train than standard frame-based approaches. While some segmental models have been successfully trained end to end, there is a lack of understanding of their training under different settings and with different losses.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121022246","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 : 2016-10-20DOI: 10.1109/SLT.2016.7846248
Shubham Toshniwal, Karen Livescu
We propose an attention-enabled encoder-decoder model for the problem of grapheme-to-phoneme conversion. Most previous work has tackled the problem via joint sequence models that require explicit alignments for training. In contrast, the attention-enabled encoder-decoder model allows for jointly learning to align and convert characters to phonemes. We explore different types of attention models, including global and local attention, and our best models achieve state-of-the-art results on three standard data sets (CMU-Dict, Pronlex, and NetTalk).
{"title":"Jointly learning to align and convert graphemes to phonemes with neural attention models","authors":"Shubham Toshniwal, Karen Livescu","doi":"10.1109/SLT.2016.7846248","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846248","url":null,"abstract":"We propose an attention-enabled encoder-decoder model for the problem of grapheme-to-phoneme conversion. Most previous work has tackled the problem via joint sequence models that require explicit alignments for training. In contrast, the attention-enabled encoder-decoder model allows for jointly learning to align and convert characters to phonemes. We explore different types of attention models, including global and local attention, and our best models achieve state-of-the-art results on three standard data sets (CMU-Dict, Pronlex, and NetTalk).","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123957625","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 : 2016-10-02DOI: 10.1109/SLT.2016.7846307
Y. Qian, P. Woodland
This paper describes the extension and optimisation of our previous work on very deep convolutional neural networks (CNNs) for effective recognition of noisy speech in the Aurora 4 task. The appropriate number of convolutional layers, the sizes of the filters, pooling operations and input feature maps are all modified: the filter and pooling sizes are reduced and dimensions of input feature maps are extended to allow adding more convolutional layers. Furthermore appropriate input padding and input feature map selection strategies are developed. In addition, an adaptation framework using joint training of very deep CNN with auxiliary features i-vector and fMLLR features is developed. These modifications give substantial word error rate reductions over the standard CNN used as baseline. Finally the very deep CNN is combined with an LSTM-RNN acoustic model and it is shown that state-level weighted log likelihood score combination in a joint acoustic model decoding scheme is very effective. On the Aurora 4 task, the very deep CNN achieves a WER of 8.81%, further 7.99% with auxiliary feature joint training, and 7.09% with LSTM-RNN joint decoding.
{"title":"Very deep convolutional neural networks for robust speech recognition","authors":"Y. Qian, P. Woodland","doi":"10.1109/SLT.2016.7846307","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846307","url":null,"abstract":"This paper describes the extension and optimisation of our previous work on very deep convolutional neural networks (CNNs) for effective recognition of noisy speech in the Aurora 4 task. The appropriate number of convolutional layers, the sizes of the filters, pooling operations and input feature maps are all modified: the filter and pooling sizes are reduced and dimensions of input feature maps are extended to allow adding more convolutional layers. Furthermore appropriate input padding and input feature map selection strategies are developed. In addition, an adaptation framework using joint training of very deep CNN with auxiliary features i-vector and fMLLR features is developed. These modifications give substantial word error rate reductions over the standard CNN used as baseline. Finally the very deep CNN is combined with an LSTM-RNN acoustic model and it is shown that state-level weighted log likelihood score combination in a joint acoustic model decoding scheme is very effective. On the Aurora 4 task, the very deep CNN achieves a WER of 8.81%, further 7.99% with auxiliary feature joint training, and 7.09% with LSTM-RNN joint decoding.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117118532","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 : 2016-09-27DOI: 10.1109/SLT.2016.7846296
Franck Dernoncourt, Ji Young Lee
Systems based on artificial neural networks (ANNs) have achieved state-of-the-art results in many natural language processing tasks. Although ANNs do not require manually engineered features, ANNs have many hyperparameters to be optimized. The choice of hyperparameters significantly impacts models' performances. However, the ANN hyperparameters are typically chosen by manual, grid, or random search, which either requires expert experiences or is computationally expensive. Recent approaches based on Bayesian optimization using Gaussian processes (GPs) is a more systematic way to automatically pinpoint optimal or near-optimal machine learning hyperparameters. Using a previously published ANN model yielding state-of-the-art results for dialog act classification, we demonstrate that optimizing hyperparameters using GP further improves the results, and reduces the computational time by a factor of 4 compared to a random search. Therefore it is a useful technique for tuning ANN models to yield the best performances for NLP tasks.
{"title":"Optimizing neural network hyperparameters with Gaussian processes for dialog act classification","authors":"Franck Dernoncourt, Ji Young Lee","doi":"10.1109/SLT.2016.7846296","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846296","url":null,"abstract":"Systems based on artificial neural networks (ANNs) have achieved state-of-the-art results in many natural language processing tasks. Although ANNs do not require manually engineered features, ANNs have many hyperparameters to be optimized. The choice of hyperparameters significantly impacts models' performances. However, the ANN hyperparameters are typically chosen by manual, grid, or random search, which either requires expert experiences or is computationally expensive. Recent approaches based on Bayesian optimization using Gaussian processes (GPs) is a more systematic way to automatically pinpoint optimal or near-optimal machine learning hyperparameters. Using a previously published ANN model yielding state-of-the-art results for dialog act classification, we demonstrate that optimizing hyperparameters using GP further improves the results, and reduces the computational time by a factor of 4 compared to a random search. Therefore it is a useful technique for tuning ANN models to yield the best performances for NLP tasks.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132930805","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 : 2016-09-26DOI: 10.1109/SLT.2016.7846283
Harishchandra Dubey, A. Sangwan, J. Hansen
Peer-Led Team Learning (PLTL) is a structured learning model where a team leader is appointed to facilitate collaborative problem solving among students for Science, Technology, Engineering and Mathematics (STEM) courses. This paper presents an informed HMM-based speaker diarization system. The minimum duration of short conversational-turns and number of participating students were fed as side information to the HMM system. A modified form of Bayesian Information Criterion (BIC) was used for iterative merging and re-segmentation. Finally, we used the diarization output to compute a novel dominance score based on unsupervised acoustic analysis.
{"title":"A robust diarization system for measuring dominance in Peer-Led Team Learning groups","authors":"Harishchandra Dubey, A. Sangwan, J. Hansen","doi":"10.1109/SLT.2016.7846283","DOIUrl":"https://doi.org/10.1109/SLT.2016.7846283","url":null,"abstract":"Peer-Led Team Learning (PLTL) is a structured learning model where a team leader is appointed to facilitate collaborative problem solving among students for Science, Technology, Engineering and Mathematics (STEM) courses. This paper presents an informed HMM-based speaker diarization system. The minimum duration of short conversational-turns and number of participating students were fed as side information to the HMM system. A modified form of Bayesian Information Criterion (BIC) was used for iterative merging and re-segmentation. Finally, we used the diarization output to compute a novel dominance score based on unsupervised acoustic analysis.","PeriodicalId":281635,"journal":{"name":"2016 IEEE Spoken Language Technology Workshop (SLT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129573308","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：