Automatic detection of syllable boundaries combining the advantages of treebank and bracketed corpora training

Müller, Karin

doi:10.3115/1073012.1073065

Cited by 15 publications

(23 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Van den Bosch (1997) reports a word error rate of 2.22% on English syllabification using inductive learning. Due to the feature "cluster size", which was not used by Müller (2001a), we are able to give an extensive qualitative evalu-ation of syllable structure considering syllable positions, as well as the complexities of consonant clusters, and the position of a consonant within a cluster.…”

Section: Discussionmentioning

confidence: 99%

“…As phoneme set, we used the symbols from the English and German SAMPA alphabet (Wells, 1997). In contrast to Müller (2001a), we did not investigate smaller training corpora, since we are interested in maximal phonological knowledge about internal word structure. Third, we train the phonological context-free grammar on the training corpus using the supervised method presented in Section 2.…”

Section: Experiments With German Datamentioning

confidence: 99%

“…We train the context-free grammar for German on the transcribed and syllabified training corpus with a simple supervised training method (Müller, 2001a). The main idea of the training method is that after a grammar transformation step, the grammar together with a parser can predict syllable boundaries of unknown phoneme strings.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Probabilistic context-free grammars for phonology

Müller

2002

Proceedings of the ACL-02 Workshop on Morphological and Phonological Learning -

Self Cite

View full text Add to dashboard Cite

We present a phonological probabilistic contextfree grammar, which describes the word and syllable structure of German words. The grammar is trained on a large corpus by a simple supervised method, and evaluated on a syllabification task achieving 96.88% word accuracy on word tokens, and 90.33% on word types. We added rules for English phonemes to the grammar, and trained the enriched grammar on an English corpus. Both grammars are evaluated qualitatively showing that probabilistic context-free grammars can contribute linguistic knowledge to phonology. Our formal approach is multilingual, while the training data is language-dependent.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Experiments With German Datamentioning

confidence: 99%

See 1 more Smart Citation

Probabilistic context-free grammars for phonology

Müller

2002

Proceedings of the ACL-02 Workshop on Morphological and Phonological Learning -

Self Cite

View full text Add to dashboard Cite

show abstract

“…With respect to either of these two model classes, each way of assigning syllable boundaries to a word corresponds to exactly one parse of that word. This makes it simple to train the models from a corpus in which syllable boundaries are provided, as in Müller (2001). We used two different corpora for our experiments, one German (from the ECI corpus of newspaper text) and one English (from the Penn WSJ corpus).…”

Section: Statistical Parsing Of Syllable Structurementioning

confidence: 99%

Representational bias in unsupervised learning of syllable structure

Goldwater

Johnson

2005

Proceedings of the Ninth Conference on Computational Natural Language Learning - CONLL '05

View full text Add to dashboard Cite

Unsupervised learning algorithms based on Expectation Maximization (EM) are often straightforward to implement and provably converge on a local likelihood maximum. However, these algorithms often do not perform well in practice. Common wisdom holds that they yield poor results because they are overly sensitive to initial parameter values and easily get stuck in local (but not global) maxima. We present a series of experiments indicating that for the task of learning syllable structure, the initial parameter weights are not crucial. Rather, it is the choice of model class itself that makes the difference between successful and unsuccessful learning. We use a language-universal rule-based algorithm to find a good set of parameters, and then train the parameter weights using EM. We achieve word accuracy of 95.9% on German and 97.1% on English, as compared to 97.4% and 98.1% respectively for supervised training.

show abstract

“…Also for Dutch, Bouma (2003) describes a 'hyphenation' method but since the 'core rule of Dutch hyphenation is that hyphenation points fall between syllables' (p. 5), this is effectively a syllabification algorithm also. The overall system uses a hand-crafted rule-based system (see Müller 2001) which produces (all possible) candidate pronunciations. Using a simple over-generative rule-based method which is then subjected to transformationbased learning (Brill 1995) to control the errors, Bouma reports an accuracy of 99.35% hyphens correct on a test set of approximately 29,000 words after training on approximately 260,000 words.…”

Section: Introductionmentioning

confidence: 99%

Can syllabification improve pronunciation by analogy of English?

Marchand

Damper

2006

Nat. Lang. Eng.

View full text Add to dashboard Cite

In spite of difficulty in defining the syllable unequivocally, and controversy over its role in theories of spoken and written language processing, the syllable is a potentially useful unit in several practical tasks which arise in computational linguistics and speech technology. For instance, syllable structure might embody valuable information for building word models in automatic speech recognition, and concatenative speech synthesis might use syllables or demisyllables as basic units. In this paper, we first present an algorithm for determining syllable boundaries in the orthographic form of unknown words that works by analogical reasoning from a database or corpus of known syllabifications. We call this syllabification by analogy (SbA). It is similarly motivated to our existing pronunciation by analogy (PbA) which predicts pronunciations for unknown words (specified by their spellings) by inference from a dictionary of known word spellings and corresponding pronunciations. We show that including perfect (according to the corpus) syllable boundary information in the orthographic input can dramatically improve the performance of pronunciation by analogy of English words, but such information would not be available to a practical system. So we next investigate combining automatically-inferred syllabification and pronunciation in two different ways: the series model in which syllabification is followed sequentially by pronunciation generation; and the parallel model in which syllabification and pronunciation are simultaneously inferred. Unfortunately, neither improves performance over PbA without syllabification. Possible reasons for this failure are explored via an analysis of syllabification and pronunciation errors.

show abstract

Automatic detection of syllable boundaries combining the advantages of treebank and bracketed corpora training

Cited by 15 publications

References 4 publications

Probabilistic context-free grammars for phonology

Probabilistic context-free grammars for phonology

Representational bias in unsupervised learning of syllable structure

Can syllabification improve pronunciation by analogy of English?

Contact Info

Product

Resources

About