Pub Date : 2013-12-01DOI: 10.1109/ASRU.2013.6707753
Guangsen Wang, K. Sim
The data sparsity problem of context-dependent acoustic modelling in automatic speech recognition is addressed by using the decision tree state clusters as the training targets in the standard context-dependent (CD) deep neural network (DNN) systems. As a result, the CD states within a cluster cannot be distinguished during decoding. This problem, referred to as the clustering problem, is not explicitly addressed in the current literature. In this paper, we formulate the CD DNN as an instance of the canonical state modelling technique based on a set of broad phone classes to address both the data sparsity and the clustering problems. The triphone is clustered into multiple sets of shorter biphones using broad phone contexts to address the data sparsity issue. A DNN is trained to discriminate the biphones within each set. The canonical states are represented by the concatenated log posteriors of all the broad phone DNNs. Logistic regression is used to transform the canonical states into the triphone state output probability. Clustering of the regression parameters is used to reduce model complexity while still achieving unique acoustic scores for all possible triphones. The experimental results on a broadcast news transcription task reveal that the proposed regression-based CD DNN significantly outperforms the standard CD DNN. The best system provides a 2.7% absolute WER reduction compared to the best standard CD DNN system.
{"title":"Context-dependent modelling of deep neural network using logistic regression","authors":"Guangsen Wang, K. Sim","doi":"10.1109/ASRU.2013.6707753","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707753","url":null,"abstract":"The data sparsity problem of context-dependent acoustic modelling in automatic speech recognition is addressed by using the decision tree state clusters as the training targets in the standard context-dependent (CD) deep neural network (DNN) systems. As a result, the CD states within a cluster cannot be distinguished during decoding. This problem, referred to as the clustering problem, is not explicitly addressed in the current literature. In this paper, we formulate the CD DNN as an instance of the canonical state modelling technique based on a set of broad phone classes to address both the data sparsity and the clustering problems. The triphone is clustered into multiple sets of shorter biphones using broad phone contexts to address the data sparsity issue. A DNN is trained to discriminate the biphones within each set. The canonical states are represented by the concatenated log posteriors of all the broad phone DNNs. Logistic regression is used to transform the canonical states into the triphone state output probability. Clustering of the regression parameters is used to reduce model complexity while still achieving unique acoustic scores for all possible triphones. The experimental results on a broadcast news transcription task reveal that the proposed regression-based CD DNN significantly outperforms the standard CD DNN. The best system provides a 2.7% absolute WER reduction compared to the best standard CD DNN system.","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":"126464659","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.2013.6707740
Florian Metze, Zaid A. W. Sheikh, A. Waibel, Jonas Gehring, Kevin Kilgour, Quoc Bao Nguyen, V. Nguyen
Conventional wisdom in automatic speech recognition asserts that pitch information is not helpful in building speech recognizers for non-tonal languages and contributes only modestly to performance in speech recognizers for tonal languages. To maintain consistency between different systems, pitch is therefore often ignored, trading the slight performance benefits for greater system uniformity/ simplicity. In this paper, we report results that challenge this conventional approach. We present new models of tone that deliver consistent performance improvements for tonal languages (Cantonese, Vietnamese) and even modest improvements for non-tonal languages. Using neural networks for feature integration and fusion, these models achieve significant gains throughout, and provide us with system uniformity and standardization across all languages, tonal and non-tonal.
{"title":"Models of tone for tonal and non-tonal languages","authors":"Florian Metze, Zaid A. W. Sheikh, A. Waibel, Jonas Gehring, Kevin Kilgour, Quoc Bao Nguyen, V. Nguyen","doi":"10.1109/ASRU.2013.6707740","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707740","url":null,"abstract":"Conventional wisdom in automatic speech recognition asserts that pitch information is not helpful in building speech recognizers for non-tonal languages and contributes only modestly to performance in speech recognizers for tonal languages. To maintain consistency between different systems, pitch is therefore often ignored, trading the slight performance benefits for greater system uniformity/ simplicity. In this paper, we report results that challenge this conventional approach. We present new models of tone that deliver consistent performance improvements for tonal languages (Cantonese, Vietnamese) and even modest improvements for non-tonal languages. Using neural networks for feature integration and fusion, these models achieve significant gains throughout, and provide us with system uniformity and standardization across all languages, tonal and non-tonal.","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":"116825248","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.2013.6707758
H. Liao, E. McDermott, A. Senior
YouTube is a highly visited video sharing website where over one billion people watch six billion hours of video every month. Improving accessibility to these videos for the hearing impaired and for search and indexing purposes is an excellent application of automatic speech recognition. However, YouTube videos are extremely challenging for automatic speech recognition systems. Standard adapted Gaussian Mixture Model (GMM) based acoustic models can have word error rates above 50%, making this one of the most difficult reported tasks. Since 2009, YouTube has provided automatic generation of closed captions for videos detected to have English speech; the service now supports ten different languages. This paper describes recent improvements to the original system, in particular the use of owner-uploaded video transcripts to generate additional semi-supervised training data and deep neural networks acoustic models with large state inventories. Applying an “island of confidence” filtering heuristic to select useful training segments, and increasing the model size by using 44,526 context dependent states with a low-rank final layer weight matrix approximation, improved performance by about 13% relative compared to previously reported sequence trained DNN results for this task.
{"title":"Large scale deep neural network acoustic modeling with semi-supervised training data for YouTube video transcription","authors":"H. Liao, E. McDermott, A. Senior","doi":"10.1109/ASRU.2013.6707758","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707758","url":null,"abstract":"YouTube is a highly visited video sharing website where over one billion people watch six billion hours of video every month. Improving accessibility to these videos for the hearing impaired and for search and indexing purposes is an excellent application of automatic speech recognition. However, YouTube videos are extremely challenging for automatic speech recognition systems. Standard adapted Gaussian Mixture Model (GMM) based acoustic models can have word error rates above 50%, making this one of the most difficult reported tasks. Since 2009, YouTube has provided automatic generation of closed captions for videos detected to have English speech; the service now supports ten different languages. This paper describes recent improvements to the original system, in particular the use of owner-uploaded video transcripts to generate additional semi-supervised training data and deep neural networks acoustic models with large state inventories. Applying an “island of confidence” filtering heuristic to select useful training segments, and increasing the model size by using 44,526 context dependent states with a low-rank final layer weight matrix approximation, improved performance by about 13% relative compared to previously reported sequence trained DNN results for this task.","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":"121653733","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.2013.6707731
D. Karakos, R. Schwartz, S. Tsakalidis, Le Zhang, Shivesh Ranjan, Tim Ng, Roger Hsiao, G. Saikumar, I. Bulyko, L. Nguyen, J. Makhoul, F. Grézl, M. Hannemann, M. Karafiát, Igor Szöke, Karel Veselý, L. Lamel, V. Le
We present two techniques that are shown to yield improved Keyword Spotting (KWS) performance when using the ATWV/MTWV performance measures: (i) score normalization, where the scores of different keywords become commensurate with each other and they more closely correspond to the probability of being correct than raw posteriors; and (ii) system combination, where the detections of multiple systems are merged together, and their scores are interpolated with weights which are optimized using MTWV as the maximization criterion. Both score normalization and system combination approaches show that significant gains in ATWV/MTWV can be obtained, sometimes on the order of 8-10 points (absolute), in five different languages. A variant of these methods resulted in the highest performance for the official surprise language evaluation for the IARPA-funded Babel project in April 2013.
{"title":"Score normalization and system combination for improved keyword spotting","authors":"D. Karakos, R. Schwartz, S. Tsakalidis, Le Zhang, Shivesh Ranjan, Tim Ng, Roger Hsiao, G. Saikumar, I. Bulyko, L. Nguyen, J. Makhoul, F. Grézl, M. Hannemann, M. Karafiát, Igor Szöke, Karel Veselý, L. Lamel, V. Le","doi":"10.1109/ASRU.2013.6707731","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707731","url":null,"abstract":"We present two techniques that are shown to yield improved Keyword Spotting (KWS) performance when using the ATWV/MTWV performance measures: (i) score normalization, where the scores of different keywords become commensurate with each other and they more closely correspond to the probability of being correct than raw posteriors; and (ii) system combination, where the detections of multiple systems are merged together, and their scores are interpolated with weights which are optimized using MTWV as the maximization criterion. Both score normalization and system combination approaches show that significant gains in ATWV/MTWV can be obtained, sometimes on the order of 8-10 points (absolute), in five different languages. A variant of these methods resulted in the highest performance for the official surprise language evaluation for the IARPA-funded Babel project in April 2013.","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":"131058900","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.2013.6707718
João Miranda, J. Neto, A. Black
In this paper, we present a technique to use the information in multiple parallel speech streams, which are approximate translations of each other, in order to improve performance in a punctuation recovery task. We first build a phraselevel alignment of these multiple streams, using phrase tables to link the phrase pairs together. The information so collected is then used to make it more likely that sentence units are equivalent across streams. We applied this technique to a number of simultaneously interpreted speeches of the European Parliament Committees, for the recovery of the full stop, in four different languages (English, Italian, Portuguese and Spanish). We observed an average improvement in SER of 37% when compared to an existing baseline, in Portuguese and English.
{"title":"Improved punctuation recovery through combination of multiple speech streams","authors":"João Miranda, J. Neto, A. Black","doi":"10.1109/ASRU.2013.6707718","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707718","url":null,"abstract":"In this paper, we present a technique to use the information in multiple parallel speech streams, which are approximate translations of each other, in order to improve performance in a punctuation recovery task. We first build a phraselevel alignment of these multiple streams, using phrase tables to link the phrase pairs together. The information so collected is then used to make it more likely that sentence units are equivalent across streams. We applied this technique to a number of simultaneously interpreted speeches of the European Parliament Committees, for the recovery of the full stop, in four different languages (English, Italian, Portuguese and Spanish). We observed an average improvement in SER of 37% when compared to an existing baseline, in Portuguese and English.","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":"122285835","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.2013.6707773
Jahn Heymann, Oliver Walter, Reinhold Häb-Umbach, B. Raj
In this paper we present an algorithm for the unsupervised segmentation of a character or phoneme lattice into words. Using a lattice at the input rather than a single string accounts for the uncertainty of the character/phoneme recognizer about the true label sequence. An example application is the discovery of lexical units from the output of an error-prone phoneme recognizer in a zero-resource setting, where neither the lexicon nor the language model is known. Recently a Weighted Finite State Transducer (WFST) based approach has been published which we show to suffer from an issue: language model probabilities of known words are computed incorrectly. Fixing this issue leads to greatly improved precision and recall rates, however at the cost of increased computational complexity. It is therefore practical only for single input strings. To allow for a lattice input and thus for errors in the character/phoneme recognizer, we propose a computationally efficient suboptimal two-stage approach, which is shown to significantly improve the word segmentation performance compared to the earlier WFST approach.
{"title":"Unsupervised word segmentation from noisy input","authors":"Jahn Heymann, Oliver Walter, Reinhold Häb-Umbach, B. Raj","doi":"10.1109/ASRU.2013.6707773","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707773","url":null,"abstract":"In this paper we present an algorithm for the unsupervised segmentation of a character or phoneme lattice into words. Using a lattice at the input rather than a single string accounts for the uncertainty of the character/phoneme recognizer about the true label sequence. An example application is the discovery of lexical units from the output of an error-prone phoneme recognizer in a zero-resource setting, where neither the lexicon nor the language model is known. Recently a Weighted Finite State Transducer (WFST) based approach has been published which we show to suffer from an issue: language model probabilities of known words are computed incorrectly. Fixing this issue leads to greatly improved precision and recall rates, however at the cost of increased computational complexity. It is therefore practical only for single input strings. To allow for a lattice input and thus for errors in the character/phoneme recognizer, we propose a computationally efficient suboptimal two-stage approach, which is shown to significantly improve the word segmentation performance compared to the earlier WFST approach.","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":"129759280","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.2013.6707708
Jingjing Liu, Panupong Pasupat, Yining Wang, D. S. Cyphers, James R. Glass
Query understanding has been well studied in the areas of information retrieval and spoken language understanding (SLU). There are generally three layers of query understanding: domain classification, user intent detection, and semantic tagging. Classifiers can be applied to domain and intent detection in real systems, and semantic tagging (or slot filling) is commonly defined as a sequence-labeling task - mapping a sequence of words to a sequence of labels. Various statistical features (e.g., n-grams) can be extracted from annotated queries for learning label prediction models; however, linguistic characteristics of queries, such as hierarchical structures and semantic relationships, are usually neglected in the feature extraction process. In this work, we propose an approach that leverages linguistic knowledge encoded in hierarchical parse trees for query understanding. Specifically, for natural language queries, we extract a set of syntactic structural features and semantic dependency features from query parse trees to enhance inference model learning. Experiments on real natural language queries show that augmenting sequence labeling models with linguistic knowledge can improve query understanding performance in various domains.
{"title":"Query understanding enhanced by hierarchical parsing structures","authors":"Jingjing Liu, Panupong Pasupat, Yining Wang, D. S. Cyphers, James R. Glass","doi":"10.1109/ASRU.2013.6707708","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707708","url":null,"abstract":"Query understanding has been well studied in the areas of information retrieval and spoken language understanding (SLU). There are generally three layers of query understanding: domain classification, user intent detection, and semantic tagging. Classifiers can be applied to domain and intent detection in real systems, and semantic tagging (or slot filling) is commonly defined as a sequence-labeling task - mapping a sequence of words to a sequence of labels. Various statistical features (e.g., n-grams) can be extracted from annotated queries for learning label prediction models; however, linguistic characteristics of queries, such as hierarchical structures and semantic relationships, are usually neglected in the feature extraction process. In this work, we propose an approach that leverages linguistic knowledge encoded in hierarchical parse trees for query understanding. Specifically, for natural language queries, we extract a set of syntactic structural features and semantic dependency features from query parse trees to enhance inference model learning. Experiments on real natural language queries show that augmenting sequence labeling models with linguistic knowledge can improve query understanding performance in various domains.","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":"125665599","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.2013.6707707
David Nolden, R. Schlüter, H. Ney
We show that most search errors can be identified by aligning the results of a symmetric forward and backward decoding pass. Based on this knowledge, we introduce an efficient high-level decoding architecture which yields virtually no search errors, and requires virtually no manual tuning. We perform an initial forward- and backward decoding with tight initial beams, then we identify search errors, and then we recursively increment the beam sizes and perform new forward and backward decodings for erroneous intervals until no more search errors are detected. Consequently, each utterance and even each single word is decoded with the smallest beam size required to decode it correctly. On all tested systems we achieve an error rate equal or very close to classical decoding with ideally tuned beam size, but unsupervisedly without specific tuning, and at around 2 times faster runtime. An additional speedup by factor 2 can be achieved by decoding the forward and backward pass in separate threads.
{"title":"Efficient nearly error-less LVCSR decoding based on incremental forward and backward passes","authors":"David Nolden, R. Schlüter, H. Ney","doi":"10.1109/ASRU.2013.6707707","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707707","url":null,"abstract":"We show that most search errors can be identified by aligning the results of a symmetric forward and backward decoding pass. Based on this knowledge, we introduce an efficient high-level decoding architecture which yields virtually no search errors, and requires virtually no manual tuning. We perform an initial forward- and backward decoding with tight initial beams, then we identify search errors, and then we recursively increment the beam sizes and perform new forward and backward decodings for erroneous intervals until no more search errors are detected. Consequently, each utterance and even each single word is decoded with the smallest beam size required to decode it correctly. On all tested systems we achieve an error rate equal or very close to classical decoding with ideally tuned beam size, but unsupervisedly without specific tuning, and at around 2 times faster runtime. An additional speedup by factor 2 can be achieved by decoding the forward and backward pass in separate threads.","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":"114515173","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.2013.6707751
Zhiheng Huang, G. Zweig, Michael Levit, Benoît Dumoulin, Barlas Oğuz, Shawn Chang
Recurrent neural network (RNN) language models have proven to be successful to lower the perplexity and word error rate in automatic speech recognition (ASR). However, one challenge to adopt RNN language models is due to their heavy computational cost in training. In this paper, we propose two techniques to accelerate RNN training: 1) two stage class RNN and 2) parallel RNN training. In experiments on Microsoft internal short message dictation (SMD) data set, two stage class RNNs and parallel RNNs not only result in equal or lower WERs compared to original RNNs but also accelerate training by 2 and 10 times respectively. It is worth noting that two stage class RNN speedup can also be applied to test stage, which is essential to reduce the latency in real time ASR applications.
{"title":"Accelerating recurrent neural network training via two stage classes and parallelization","authors":"Zhiheng Huang, G. Zweig, Michael Levit, Benoît Dumoulin, Barlas Oğuz, Shawn Chang","doi":"10.1109/ASRU.2013.6707751","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707751","url":null,"abstract":"Recurrent neural network (RNN) language models have proven to be successful to lower the perplexity and word error rate in automatic speech recognition (ASR). However, one challenge to adopt RNN language models is due to their heavy computational cost in training. In this paper, we propose two techniques to accelerate RNN training: 1) two stage class RNN and 2) parallel RNN training. In experiments on Microsoft internal short message dictation (SMD) data set, two stage class RNNs and parallel RNNs not only result in equal or lower WERs compared to original RNNs but also accelerate training by 2 and 10 times respectively. It is worth noting that two stage class RNN speedup can also be applied to test stage, which is essential to reduce the latency in real time ASR applications.","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":"130117254","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.2013.6707766
Guoguo Chen, Oguz Yilmaz, J. Trmal, Daniel Povey, S. Khudanpur
We propose a simple but effective weighted finite state transducer (WFST) based framework for handling out-of-vocabulary (OOV) keywords in a speech search task. State-of-the-art large vocabulary continuous speech recognition (LVCSR) and keyword search (KWS) systems are developed for conversational telephone speech in Tagalog. Word-based and phone-based indexes are created from word lattices, the latter by using the LVCSR system's pronunciation lexicon. Pronunciations of OOV keywords are hypothesized via a standard grapheme-to-phoneme method. In-vocabulary proxies (word or phone sequences) are generated for each OOV keyword using WFST techniques that permit incorporation of a phone confusion matrix. Empirical results when searching for the Babel/NIST evaluation keywords in the Babel 10 hour development-test speech collection show that (i) searching for word proxies in the word index significantly outperforms searching for phonetic representations of OOV words in a phone index, and (ii) while phone confusion information yields minor improvement when searching a phone index, it yields up to 40% improvement in actual term weighted value when searching a word index with word proxies.
{"title":"Using proxies for OOV keywords in the keyword search task","authors":"Guoguo Chen, Oguz Yilmaz, J. Trmal, Daniel Povey, S. Khudanpur","doi":"10.1109/ASRU.2013.6707766","DOIUrl":"https://doi.org/10.1109/ASRU.2013.6707766","url":null,"abstract":"We propose a simple but effective weighted finite state transducer (WFST) based framework for handling out-of-vocabulary (OOV) keywords in a speech search task. State-of-the-art large vocabulary continuous speech recognition (LVCSR) and keyword search (KWS) systems are developed for conversational telephone speech in Tagalog. Word-based and phone-based indexes are created from word lattices, the latter by using the LVCSR system's pronunciation lexicon. Pronunciations of OOV keywords are hypothesized via a standard grapheme-to-phoneme method. In-vocabulary proxies (word or phone sequences) are generated for each OOV keyword using WFST techniques that permit incorporation of a phone confusion matrix. Empirical results when searching for the Babel/NIST evaluation keywords in the Babel 10 hour development-test speech collection show that (i) searching for word proxies in the word index significantly outperforms searching for phonetic representations of OOV words in a phone index, and (ii) while phone confusion information yields minor improvement when searching a phone index, it yields up to 40% improvement in actual term weighted value when searching a word index with word proxies.","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":"117304124","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}