Using closely-related language to build an ASR for a very under-resourced language: Iban

Juan, Sarah Samson; Besacier, Laurent; Lecouteux, Benjamin; Tan, Tien-Ping

doi:10.1109/icsda.2014.7051423

Cited by 5 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using language resources from high-resourced languages for the recognition of an under-resourced language is common practice [35][36][37]. Being an under-resourced language, Frisian also lacks adequate speech data to train acoustic models that can provide accurate enough recognition.…”

Section: Multilingual Trainingmentioning

confidence: 99%

Code-switching detection using multilingual DNNS

Yılmaz

Heuvel

Leeuwen

2016

2016 IEEE Spoken Language Technology Workshop (SLT)

View full text Add to dashboard Cite

Automatic speech recognition (ASR) of code-switching speech requires careful handling of unexpected language switches that may occur in a single utterance. In this paper, we investigate the feasibility of using multilingually trained deep neural networks (DNN) for the ASR of Frisian speech containing code-switches to Dutch with the aim of building a robust recognizer that can handle this phenomenon. For this purpose, we train several multilingual DNN models on Frisian and two closely related languages, namely English and Dutch, to compare the impact of single-step and two-step multilingual DNN training on the recognition and code-switching detection performance. We apply bilingual DNN retraining on both target languages by varying the amount of training data belonging to the higher-resourced target language (Dutch). The recognition results show that the multilingual DNN training scheme with an initial multilingual training step followed by bilingual retraining provides recognition performance comparable to an oracle baseline recognizer that can employ language-specific acoustic models. We further show that we can detect code-switches at the word level with an equal error rate of around 17% excluding the deletions due to ASR errors.

show abstract

Section: Multilingual Trainingmentioning

confidence: 99%

Code-switching detection using multilingual DNNS

Yılmaz

Heuvel

Leeuwen

2016

2016 IEEE Spoken Language Technology Workshop (SLT)

View full text Add to dashboard Cite

show abstract

“…We used data sets for four widely spoken lowresource languages, Fongbe (Laleye et al, 2016), Wolof (Gauthier et al, 2016), Swahili (Gelas et al, 2012), and Iban (Juan et al, 2014), which were previously released as ASR corpora. They include segmented audio with corresponding transcripts, as well as additional written texts for training the language model (see Table 1 for details).…”

Section: Data Descriptionsmentioning

confidence: 99%

Data-driven Model Generalizability in Crosslinguistic Low-resource Morphological Segmentation

Liu

Prud’hommeaux

2022

Transactions of the Association for Computational Linguistics

View full text Add to dashboard Cite

Common designs of model evaluation typically focus on monolingual settings, where different models are compared according to their performance on a single data set that is assumed to be representative of all possible data for the task at hand. While this may be reasonable for a large data set, this assumption is difficult to maintain in low-resource scenarios, where artifacts of the data collection can yield data sets that are outliers, potentially making conclusions about model performance coincidental. To address these concerns, we investigate model generalizability in crosslinguistic low-resource scenarios. Using morphological segmentation as the test case, we compare three broad classes of models with different parameterizations, taking data from 11 languages across 6 language families. In each experimental setting, we evaluate all models on a first data set, then examine their performance consistency when introducing new randomly sampled data sets with the same size and when applying the trained models to unseen test sets of varying sizes. The results demonstrate that the extent of model generalization depends on the characteristics of the data set, and does not necessarily rely heavily on the data set size. Among the characteristics that we studied, the ratio of morpheme overlap and that of the average number of morphemes per word between the training and test sets are the two most prominent factors. Our findings suggest that future work should adopt random sampling to construct data sets with different sizes in order to make more responsible claims about model evaluation.

show abstract

“…Languages were chosen from the CMU Wilderness dataset given availability of additional data in a different setting, and include several language families as well as more closelyrelated challenge pairs such as Javanese and Sundanese. These included data from the Common Voice project (CV; Ardila et al, 2020) which is read speech typically recorded using built-in laptop microphones; radio news data (SLR24; Juan et al, 2014Juan et al, , 2015; crowd-sourced recordings using portable electronics (SLR35, SLR36; Kjartansson et al, 2018); cleanly recorded microphone data (SLR64, SLR65, SLR66, SLR79; He et al, 2020); and a collection of recordings from varied sources (SS; Shukla, 2020). Table 1 shows the task languages and their data sources for evaluation splits for the robust language identification task.…”

Section: Provided Datamentioning

confidence: 99%

SIGTYP 2021 Shared Task: Robust Spoken Language Identification

Salesky¹,

Abdullah²,

Mielke³

et al. 2021

Proceedings of the Third Workshop on Computational Typology and Multilingual NLP

View full text Add to dashboard Cite

While language identification is a fundamental speech and language processing task, for many languages and language families it remains a challenging task. For many low-resource and endangered languages this is in part due to resource availability: where larger datasets exist, they may be single-speaker or have different domains than desired application scenarios, demanding a need for domain and speakerinvariant language identification systems. This year's shared task on robust spoken language identification sought to investigate just this scenario: systems were to be trained on largely single-speaker speech from one domain, but evaluated on data in other domains recorded from speakers under different recording circumstances, mimicking realistic low-resource scenarios. We see that domain and speaker mismatch proves very challenging for current methods which can perform above 95% accuracy in-domain, which domain adaptation can address to some degree, but that these conditions merit further investigation to make spoken language identification accessible in many scenarios.

show abstract

Using closely-related language to build an ASR for a very under-resourced language: Iban

Cited by 5 publications

References 18 publications

Code-switching detection using multilingual DNNS

Code-switching detection using multilingual DNNS

Data-driven Model Generalizability in Crosslinguistic Low-resource Morphological Segmentation

SIGTYP 2021 Shared Task: Robust Spoken Language Identification

Contact Info

Product

Resources

About