This paper conceptualizes speech prosody data mining and its potential application in data-driven phonology/phonetics research. We first conceptualize Speech Prosody Mining (SPM) in a time-series data mining framework. Specifically, we propose using efficient symbolic representations for speech prosody time-series similarity computation. We experiment with both symbolic and numeric representations and distance measures in a series of time-series classification and clustering experiments on a dataset of Mandarin tones. Evaluation results show that symbolic representation performs comparably with other representations at a reduced cost, which enables us to efficiently mine large speech prosody corpora while opening up to possibilities of using a wide range of algorithms that require discrete valued data. We discuss the potential of SPM using time-series mining techniques in future works.
{"title":"Mining linguistic tone patterns with symbolic representation","authors":"Shuo Zhang","doi":"10.18653/v1/W16-2001","DOIUrl":"https://doi.org/10.18653/v1/W16-2001","url":null,"abstract":"This paper conceptualizes speech prosody data mining and its potential application in data-driven phonology/phonetics research. We first conceptualize Speech Prosody Mining (SPM) in a time-series data mining framework. Specifically, we propose using efficient symbolic representations for speech prosody time-series similarity computation. We experiment with both symbolic and numeric representations and distance measures in a series of time-series classification and clustering experiments on a dataset of Mandarin tones. Evaluation results show that symbolic representation performs comparably with other representations at a reduced cost, which enables us to efficiently mine large speech prosody corpora while opening up to possibilities of using a wide range of algorithms that require discrete valued data. We discuss the potential of SPM using time-series mining techniques in future works.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122768288","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}
Philippa Shoemark, S. Goldwater, James P. Kirby, Rik Sarkar
Recent work has proposed using network science to analyse the structure of the mental lexicon by viewing words as nodes in a phonological network, with edges connecting words that differ by a single phoneme. Comparing the structure of phonological networks across different languages could provide insights into linguistic typology and the cognitive pressures that shape language acquisition, evolution, and processing. However, previous studies have not considered how statistics gathered from these networks are affected by factors such as lexicon size and the distribution of word lengths. We show that these factors can substantially affect the statistics of a phonological network and propose a new method for making more robust comparisons. We then analyse eight languages, finding many commonalities but also some qualitative differences in their lexicon structure.
{"title":"Towards robust cross-linguistic comparisons of phonological networks","authors":"Philippa Shoemark, S. Goldwater, James P. Kirby, Rik Sarkar","doi":"10.18653/v1/W16-2018","DOIUrl":"https://doi.org/10.18653/v1/W16-2018","url":null,"abstract":"Recent work has proposed using network science to analyse the structure of the mental lexicon by viewing words as nodes in a phonological network, with edges connecting words that differ by a single phoneme. Comparing the structure of phonological networks across different languages could provide insights into linguistic typology and the cognitive pressures that shape language acquisition, evolution, and processing. However, previous studies have not considered how statistics gathered from these networks are affected by factors such as lexicon size and the distribution of word lengths. We show that these factors can substantially affect the statistics of a phonological network and propose a new method for making more robust comparisons. We then analyse eight languages, finding many commonalities but also some qualitative differences in their lexicon structure.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121700364","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}
Ryan Cotterell, Christo Kirov, John Sylak-Glassman, David Yarowsky, Jason Eisner, Mans Hulden
The 2016 SIGMORPHON Shared Task was devoted to the problem of morphological reinflection. It introduced morphological datasets for 10 languages with diverse ty-pological characteristics. The shared task drew submissions from 9 teams representing 11 institutions reflecting a variety of approaches to addressing supervised learning of reinflection. For the simplest task, in-flection generation from lemmas, the best system averaged 95.56% exact-match accuracy across all languages, ranging from Maltese (88.99%) to Hungarian (99.30%). With the relatively large training datasets provided, recurrent neural network architectures consistently performed best—in fact, there was a significant margin between neural and non-neural approaches. The best neural approach, averaged over all tasks and languages, outperformed the best non-neural one by 13.76% absolute; on individual tasks and languages the gap in accuracy sometimes exceeded 60%. Overall, the results show a strong state of the art, and serve as encouragement for future shared tasks that explore morphological analysis and generation with varying degrees of supervision.
{"title":"The SIGMORPHON 2016 Shared Task—Morphological Reinflection","authors":"Ryan Cotterell, Christo Kirov, John Sylak-Glassman, David Yarowsky, Jason Eisner, Mans Hulden","doi":"10.18653/v1/W16-2002","DOIUrl":"https://doi.org/10.18653/v1/W16-2002","url":null,"abstract":"The 2016 SIGMORPHON Shared Task was devoted to the problem of morphological reinflection. It introduced morphological datasets for 10 languages with diverse ty-pological characteristics. The shared task drew submissions from 9 teams representing 11 institutions reflecting a variety of approaches to addressing supervised learning of reinflection. For the simplest task, in-flection generation from lemmas, the best system averaged 95.56% exact-match accuracy across all languages, ranging from Maltese (88.99%) to Hungarian (99.30%). With the relatively large training datasets provided, recurrent neural network architectures consistently performed best—in fact, there was a significant margin between neural and non-neural approaches. The best neural approach, averaged over all tasks and languages, outperformed the best non-neural one by 13.76% absolute; on individual tasks and languages the gap in accuracy sometimes exceeded 60%. Overall, the results show a strong state of the art, and serve as encouragement for future shared tasks that explore morphological analysis and generation with varying degrees of supervision.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116510257","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}
Three popular vocal-tract animation paradigms were tested for intelligibility when displaying videos of pre-recorded Electromagnetic Articulography (EMA) data in an online experiment. EMA tracks the position of sensors attached to the tongue. The conditions were dots with tails (where only the coil location is presented), 2D animation (where the dots are connected to form 2D representations of the lips, tongue surface and chin), and a 3D model with coil locations driving facial and tongue rigs. The 2D animation (recorded in VisArtico) showed the highest identification of the prompts.
{"title":"Read my points: Effect of animation type when speech-reading from EMA data","authors":"Kristy James, Martijn B. Wieling","doi":"10.18653/v1/W16-2014","DOIUrl":"https://doi.org/10.18653/v1/W16-2014","url":null,"abstract":"Three popular vocal-tract animation paradigms were tested for intelligibility when displaying videos of pre-recorded Electromagnetic Articulography (EMA) data in an online experiment. EMA tracks the position of sensors attached to the tongue. The conditions were dots with tails (where only the coil location is presented), 2D animation (where the dots are connected to form 2D representations of the lips, tongue surface and chin), and a 3D model with coil locations driving facial and tongue rigs. The 2D animation (recorded in VisArtico) showed the highest identification of the prompts.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131978279","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}
We present a high-level description and error analysis of the Columbia-NYUAD sys-tem for morphological reinflection, which builds on previous work on supervised morphological paradigm completion. Our system improved over the shared task baseline on some of the languages, reaching up to 30% absolute increase. Our ranking on average was 5th in Track 1, 8th in Track 2, and 3rd in Track 3.
{"title":"The Columbia University - New York University Abu Dhabi SIGMORPHON 2016 Morphological Reinflection Shared Task Submission","authors":"Dima Taji, R. Eskander, Nizar Habash, Owen Rambow","doi":"10.18653/v1/W16-2011","DOIUrl":"https://doi.org/10.18653/v1/W16-2011","url":null,"abstract":"We present a high-level description and error analysis of the Columbia-NYUAD sys-tem for morphological reinflection, which builds on previous work on supervised morphological paradigm completion. Our system improved over the shared task baseline on some of the languages, reaching up to 30% absolute increase. Our ranking on average was 5th in Track 1, 8th in Track 2, and 3rd in Track 3.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116884907","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}
In this paper I present a k-means clustering approach to inferring morphological position classes (morphotactics) from Interlinear Glossed Text (IGT), data collections available for some endangered and low-resource languages. While the experiment is not restricted to low-resource languages, they are meant to be the targeted domain. Specifically my approach is meant to be for field linguists who do not necessarily know how many position classes there are in the language they work with and what the position classes are, but have the expertise to evaluate different hypotheses. It builds on an existing approach (Wax, 2014), but replaces the core heuristic with a clustering algorithm. The results mainly illustrate two points. First, they are largely negative, which shows that the baseline algorithm (summarized in the paper) uses a very predictive feature to determine whether affixes belong to the same position class, namely edge overlap in the affix graph. At the same time, unlike the baseline method that relies entirely on a single feature, kmeans clustering can account for different features and helps discover more morphological phenomena, e.g. circumfixation. I conclude that unsupervised learning algorithms such as k-means clustering can in principle be used for morphotactics inference, though the algorithm should probably weigh certain features more than others. Most importantly, I conclude that clustering is a promising approach for diverse morphotactics and as such it can facilitate linguistic analysis of field languages.
{"title":"Inferring Morphotactics from Interlinear Glossed Text: Combining Clustering and Precision Grammars","authors":"Olga Zamaraeva","doi":"10.18653/v1/W16-2021","DOIUrl":"https://doi.org/10.18653/v1/W16-2021","url":null,"abstract":"In this paper I present a k-means clustering approach to inferring morphological position classes (morphotactics) from Interlinear Glossed Text (IGT), data collections available for some endangered and low-resource languages. While the experiment is not restricted to low-resource languages, they are meant to be the targeted domain. Specifically my approach is meant to be for field linguists who do not necessarily know how many position classes there are in the language they work with and what the position classes are, but have the expertise to evaluate different hypotheses. It builds on an existing approach (Wax, 2014), but replaces the core heuristic with a clustering algorithm. The results mainly illustrate two points. First, they are largely negative, which shows that the baseline algorithm (summarized in the paper) uses a very predictive feature to determine whether affixes belong to the same position class, namely edge overlap in the affix graph. At the same time, unlike the baseline method that relies entirely on a single feature, kmeans clustering can account for different features and helps discover more morphological phenomena, e.g. circumfixation. I conclude that unsupervised learning algorithms such as k-means clustering can in principle be used for morphotactics inference, though the algorithm should probably weigh certain features more than others. Most importantly, I conclude that clustering is a promising approach for diverse morphotactics and as such it can facilitate linguistic analysis of field languages.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115940751","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}
This paper describes our participation in the SIGMORPHON 2016 shared task on morphological reinflection. In the task, we use a linear-chain conditional random field model to learn to map sequences of input characters to sequences of output characters and focus on developing features that are useful for predicting inflectional behavior. Since the training data in the task is limited, we also generalize the training data by extracting, in an unsupervised fashion, the types of consonant-vowel sequences that trigger inflectional behavior, and by extending the available training data through inference of unlabeled morphosyntactic descriptions.
{"title":"Morphological reinflection with conditional random fields and unsupervised features","authors":"L. Liu, L. Mao","doi":"10.18653/v1/W16-2006","DOIUrl":"https://doi.org/10.18653/v1/W16-2006","url":null,"abstract":"This paper describes our participation in the SIGMORPHON 2016 shared task on morphological reinflection. In the task, we use a linear-chain conditional random field model to learn to map sequences of input characters to sequences of output characters and focus on developing features that are useful for predicting inflectional behavior. Since the training data in the task is limited, we also generalize the training data by extracting, in an unsupervised fashion, the types of consonant-vowel sequences that trigger inflectional behavior, and by extending the available training data through inference of unlabeled morphosyntactic descriptions.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115647634","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}
M. Kisselew, Laura Rimell, Alexis Palmer, Sebastian Padó
Conversion is a word formation operation that changes the grammatical category of a word in the absence of overt morphology. Conversion is extremely productive in English (e.g., tunnel, talk). This paper investigates whether distributional information can be used to predict the diachronic direction of conversion for homophonous noun‐verb pairs. We aim to predict, for example, that tunnel was used as a noun prior to its use as a verb. We test two hypotheses: (1) that derived forms are less frequent than their bases, and (2) that derived forms are more semantically specific than their bases, as approximated by information theoretic measures. We find that hypothesis (1) holds for N-to-V conversion, while hypothesis (2) holds for V-to-N conversion. We achieve the best overall account of the historical data by taking both frequency and semantic specificity into account. These results provide a new perspective on linguistic theories regarding the semantic specificity of derivational morphemes, and on the morphosyntactic status of conversion.
{"title":"Predicting the Direction of Derivation in English Conversion","authors":"M. Kisselew, Laura Rimell, Alexis Palmer, Sebastian Padó","doi":"10.18653/v1/W16-2015","DOIUrl":"https://doi.org/10.18653/v1/W16-2015","url":null,"abstract":"Conversion is a word formation operation that changes the grammatical category of a word in the absence of overt morphology. Conversion is extremely productive in English (e.g., tunnel, talk). This paper investigates whether distributional information can be used to predict the diachronic direction of conversion for homophonous noun‐verb pairs. We aim to predict, for example, that tunnel was used as a noun prior to its use as a verb. We test two hypotheses: (1) that derived forms are less frequent than their bases, and (2) that derived forms are more semantically specific than their bases, as approximated by information theoretic measures. We find that hypothesis (1) holds for N-to-V conversion, while hypothesis (2) holds for V-to-N conversion. We achieve the best overall account of the historical data by taking both frequency and semantic specificity into account. These results provide a new perspective on linguistic theories regarding the semantic specificity of derivational morphemes, and on the morphosyntactic status of conversion.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122935832","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}
This paper presents a proposal for learning morphological inflections by a graphemeto-phoneme learning model. No special processing is used for specific languages. The starting point has been our previous research on induction of phonology and morphology for normalization of historical texts. The results show that a very simple method can indeed improve upon some baselines, but does not reach the accuracies of the best systems in the task.
{"title":"EHU at the SIGMORPHON 2016 Shared Task. A Simple Proposal: Grapheme-to-Phoneme for Inflection","authors":"I. Alegria, Izaskun Etxeberria","doi":"10.18653/v1/W16-2004","DOIUrl":"https://doi.org/10.18653/v1/W16-2004","url":null,"abstract":"This paper presents a proposal for learning morphological inflections by a graphemeto-phoneme learning model. No special processing is used for specific languages. The starting point has been our previous research on induction of phonology and morphology for normalization of historical texts. The results show that a very simple method can indeed improve upon some baselines, but does not reach the accuracies of the best systems in the task.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127851603","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}
We examine the typology of quantity-insensitive (QI) stress systems and ask to what extent an existing optimality theoretic model of QI stress can predict the observed typological frequencies of stress patterns. We find three significant correlates of pattern attestation and frequency: the trigram entropy of a pattern, the degree to which it is "confusable" with other patterns predicted by the model, and the number of constraint rankings that specify the pattern.
{"title":"Three Correlates of the Typological Frequency of Quantity-Insensitive Stress Systems","authors":"Max Bane, Jason Riggle","doi":"10.3115/1626324.1626330","DOIUrl":"https://doi.org/10.3115/1626324.1626330","url":null,"abstract":"We examine the typology of quantity-insensitive (QI) stress systems and ask to what extent an existing optimality theoretic model of QI stress can predict the observed typological frequencies of stress patterns. We find three significant correlates of pattern attestation and frequency: the trigram entropy of a pattern, the degree to which it is \"confusable\" with other patterns predicted by the model, and the number of constraint rankings that specify the pattern.","PeriodicalId":186158,"journal":{"name":"Special Interest Group on Computational Morphology and Phonology Workshop","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132272855","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：