Developing Pronunciation Models in New Languages Faster by Exploiting Common Grapheme-to-Phoneme Correspondences Across Languages

Bleyan, Harry; Ritchie, Sandy; Mortensen, Jonas Fromseier; Esch, Daan van

doi:10.21437/interspeech.2019-1781

Cited by 3 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Only a limited number of papers so far focus on developing G2P models for unseen languages. The most common strategy is to drop the target language information and make predictions using a shared multilingual model (Peters et al, 2017;Bleyan et al, 2019). This is one of our baseline (the global language model) in this work.…”

Section: Related Workmentioning

confidence: 99%

Zero-shot Learning for Grapheme to Phoneme Conversion with Language Ensemble

Li¹,

Metze²,

Mortensen³

et al. 2022

Findings of the Association for Computational Linguistics: ACL 2022

View full text Add to dashboard Cite

Grapheme-to-Phoneme (G2P) has many applications in NLP and speech fields. Most existing work focuses heavily on languages with abundant training datasets, which limits the scope of target languages to less than 100 languages. This work attempts to apply zero-shot learning to approximate G2P models for all lowresource and endangered languages in Glottolog (about 8k languages). For any unseen target language, we first build the phylogenetic tree (i.e. language family tree) to identify top-k nearest languages for which we have training sets. Then we run models of those languages to obtain a hypothesis set, which we combine into a confusion network to propose a most likely hypothesis as an approximation to the target language. We test our approach on over 600 unseen languages and demonstrate it significantly outperforms baselines. 1

show abstract

Section: Related Workmentioning

confidence: 99%

Zero-shot Learning for Grapheme to Phoneme Conversion with Language Ensemble

Li¹,

Metze²,

Mortensen³

et al. 2022

Findings of the Association for Computational Linguistics: ACL 2022

View full text Add to dashboard Cite

show abstract

“…If there is a G2P mapping and some text available in a low-resource language, an ASR system can be built by repurposing an acoustic model from a similar higher-resource language (Prasad et al, 2019). G2P mappings can even be bootstrapped for new languages using only the language's phoneme inventory combined with higherresource language text written in the same script (Bleyan et al, 2019), making even simple phoneme data a more useful tool for scaling technologies to new languages.…”

Section: Uses For Pronunciation Datamentioning

confidence: 99%

Mining Large-Scale Low-Resource Pronunciation Data From Wikipedia

Chakraborty,

Prasad,

Breiner

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Pronunciation modeling is a key task for building speech technology in new languages, and while solid grapheme-to-phoneme (G2P) mapping systems exist, language coverage can stand to be improved. The information needed to build G2P models for many more languages can easily be found on Wikipedia, but unfortunately, it is stored in disparate formats. We report on a system we built to mine a pronunciation data set in 819 languages from loosely structured tables within Wikipedia. The data includes phoneme inventories, and for 63 low-resource languages, also includes the grapheme-to-phoneme (G2P) mapping. 54 of these languages do not have easily findable G2P mappings online otherwise. We turned the information from Wikipedia into a structured, machine-readable TSV format, and make the resulting data set publicly available so it can be improved further and used in a variety of applications involving low-resource languages.

show abstract

“…For ASR it is possible to combine a target-language language model with an acoustic model from a phonologically similar language, with no need for parallel datasets of audio recordings and transcriptions . Such approaches are likely to get even more effective with nearly-universal acoustic models (Li et al, 2020) and more scalable grapheme-to-phoneme modeling approaches (Deri and Knight, 2016;Mortensen et al, 2018;Bleyan et al, 2019;Ritchie et al, 2020;. Even if more work is needed to establish when such approaches will work well (Marchisio et al, 2020;Artetxe et al, 2020;Wu and Dredze, 2020), having useful monolingual text corpora across languages is clearly a prerequisite to exploring such approaches further.…”

Section: Introductionmentioning

confidence: 99%

Language ID in the Wild: Unexpected Challenges on the Path to a Thousand-Language Web Text Corpus

Caswell

Breiner

Esch

et al. 2020

Proceedings of the 28th International Conference on Computational Linguistics

View full text Add to dashboard Cite

Large text corpora are increasingly important for a wide variety of Natural Language Processing (NLP) tasks, and automatic language identification (LangID) is a core technology needed to collect such datasets in a multilingual context. LangID is largely treated as solved in the literature, with models reported that achieve over 90% average F1 on as many as 1,366 languages. We train LangID models on up to 1,629 languages with comparable quality on held-out test sets, but find that human-judged LangID accuracy for web-crawl text corpora created using these models is only around 5% for many lower-resource languages, suggesting a need for more robust evaluation. Further analysis revealed a variety of error modes, arising from domain mismatch, class imbalance, language similarity, and insufficiently expressive models. We propose two classes of techniques to mitigate these errors: wordlist-based tunable-precision filters (for which we release curated lists in about 500 languages) and transformer-based semi-supervised LangID models, which increase median dataset precision from 5.5% to 71.2%. These techniques enable us to create an initial data set covering 100K or more relatively clean sentences in each of 500+ languages, paving the way towards a 1,000-language web text corpus.

show abstract

Developing Pronunciation Models in New Languages Faster by Exploiting Common Grapheme-to-Phoneme Correspondences Across Languages

Cited by 3 publications

References 9 publications

Zero-shot Learning for Grapheme to Phoneme Conversion with Language Ensemble

Zero-shot Learning for Grapheme to Phoneme Conversion with Language Ensemble

Mining Large-Scale Low-Resource Pronunciation Data From Wikipedia

Language ID in the Wild: Unexpected Challenges on the Path to a Thousand-Language Web Text Corpus

Contact Info

Product

Resources

About