Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case
Recently, there has been a growth in the number of studies addressing the automatic processing of low-resource languages. The lack of speech and text data significantly hinders the development of speech technologies for such languages. This paper introduces an automatic speech recognition system for...
Published in: | Mathematics |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2023
|
Subjects: | |
Online Access: | https://doi.org/10.3390/math11183814 https://doaj.org/article/d9e0c8bc649d4be88c9f93a1bca08ef9 |
id |
ftdoajarticles:oai:doaj.org/article:d9e0c8bc649d4be88c9f93a1bca08ef9 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:d9e0c8bc649d4be88c9f93a1bca08ef9 2023-10-29T02:37:35+01:00 Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case Irina Kipyatkova Ildar Kagirov 2023-09-01T00:00:00Z https://doi.org/10.3390/math11183814 https://doaj.org/article/d9e0c8bc649d4be88c9f93a1bca08ef9 EN eng MDPI AG https://www.mdpi.com/2227-7390/11/18/3814 https://doaj.org/toc/2227-7390 doi:10.3390/math11183814 2227-7390 https://doaj.org/article/d9e0c8bc649d4be88c9f93a1bca08ef9 Mathematics, Vol 11, Iss 3814, p 3814 (2023) low-resource languages automatic speech recognition audio data augmentation time delay neural network hidden Markov models long short-term memory Mathematics QA1-939 article 2023 ftdoajarticles https://doi.org/10.3390/math11183814 2023-10-01T00:37:42Z Recently, there has been a growth in the number of studies addressing the automatic processing of low-resource languages. The lack of speech and text data significantly hinders the development of speech technologies for such languages. This paper introduces an automatic speech recognition system for Livvi-Karelian. Acoustic models based on artificial neural networks with time delays and hidden Markov models were trained using a limited speech dataset of 3.5 h. To augment the data, pitch and speech rate perturbation, SpecAugment, and their combinations were employed. Language models based on 3-grams and neural networks were trained using written texts and transcripts. The achieved word error rate metric of 22.80% is comparable to other low-resource languages. To the best of our knowledge, this is the first speech recognition system for Livvi-Karelian. The results obtained can be of a certain significance for development of automatic speech recognition systems not only for Livvi-Karelian, but also for other low-resource languages, including the fields of speech recognition and machine translation systems. Future work includes experiments with Karelian data using techniques such as transfer learning and DNN language models. Article in Journal/Newspaper karelian Directory of Open Access Journals: DOAJ Articles Mathematics 11 18 3814 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
low-resource languages automatic speech recognition audio data augmentation time delay neural network hidden Markov models long short-term memory Mathematics QA1-939 |
spellingShingle |
low-resource languages automatic speech recognition audio data augmentation time delay neural network hidden Markov models long short-term memory Mathematics QA1-939 Irina Kipyatkova Ildar Kagirov Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case |
topic_facet |
low-resource languages automatic speech recognition audio data augmentation time delay neural network hidden Markov models long short-term memory Mathematics QA1-939 |
description |
Recently, there has been a growth in the number of studies addressing the automatic processing of low-resource languages. The lack of speech and text data significantly hinders the development of speech technologies for such languages. This paper introduces an automatic speech recognition system for Livvi-Karelian. Acoustic models based on artificial neural networks with time delays and hidden Markov models were trained using a limited speech dataset of 3.5 h. To augment the data, pitch and speech rate perturbation, SpecAugment, and their combinations were employed. Language models based on 3-grams and neural networks were trained using written texts and transcripts. The achieved word error rate metric of 22.80% is comparable to other low-resource languages. To the best of our knowledge, this is the first speech recognition system for Livvi-Karelian. The results obtained can be of a certain significance for development of automatic speech recognition systems not only for Livvi-Karelian, but also for other low-resource languages, including the fields of speech recognition and machine translation systems. Future work includes experiments with Karelian data using techniques such as transfer learning and DNN language models. |
format |
Article in Journal/Newspaper |
author |
Irina Kipyatkova Ildar Kagirov |
author_facet |
Irina Kipyatkova Ildar Kagirov |
author_sort |
Irina Kipyatkova |
title |
Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case |
title_short |
Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case |
title_full |
Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case |
title_fullStr |
Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case |
title_full_unstemmed |
Deep Models for Low-Resourced Speech Recognition: Livvi-Karelian Case |
title_sort |
deep models for low-resourced speech recognition: livvi-karelian case |
publisher |
MDPI AG |
publishDate |
2023 |
url |
https://doi.org/10.3390/math11183814 https://doaj.org/article/d9e0c8bc649d4be88c9f93a1bca08ef9 |
genre |
karelian |
genre_facet |
karelian |
op_source |
Mathematics, Vol 11, Iss 3814, p 3814 (2023) |
op_relation |
https://www.mdpi.com/2227-7390/11/18/3814 https://doaj.org/toc/2227-7390 doi:10.3390/math11183814 2227-7390 https://doaj.org/article/d9e0c8bc649d4be88c9f93a1bca08ef9 |
op_doi |
https://doi.org/10.3390/math11183814 |
container_title |
Mathematics |
container_volume |
11 |
container_issue |
18 |
container_start_page |
3814 |
_version_ |
1781062518240706560 |