Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic
Autor(a) principal: | |
---|---|
Data de Publicação: | 2024 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Texto livre |
Texto Completo: | https://periodicos.ufmg.br/index.php/textolivre/article/view/46952 |
Resumo: | This paper examines the errors that the automatic speech recognition (ASR) system of Veed.io produces when transcribing utterances spoken in Jordanian Arabic into subtitles. It attempts to propose a new classification for the subtitles that are built based on artificial intelligence technology. Through a combination of qualitative and quantitative analyses, the study examines the types of errors and their impact on comprehension. The errors observed in the generated subtitles based on linguistic and phonetic analysis are categorised into three main types: deletions, substitutions, and insertions. Furthermore, the quantitative analysis measures the word error rate (WER) and shows that the WER percentage is 38.857% revealing that deletions are the most common type of error, followed by substitutions and insertions. The study recommends conducting further research on ASR systems for Arabic language dialects and advises subtitlers to be aware of the limitations of these systems when using them, ensuring that they edit and supervise them appropriately. |
id |
UFMG-9_9c8a583743b1112ef1261b5e106c739c |
---|---|
oai_identifier_str |
oai:periodicos.ufmg.br:article/46952 |
network_acronym_str |
UFMG-9 |
network_name_str |
Texto livre |
repository_id_str |
|
spelling |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian ArabicLegendas em árabe geradas por inteligência artificial: insights do sistema de reconhecimento automático de fala do árabe jordaniano da Veed.ioSubtitlesAuto-generated subtitlesAutomatic Speech RecognitionLinguisticsJordanian ArabicLegendasLegendas geradas automaticamenteReconhecimento Automático de FalaLinguísticaÁrabe jordanianoThis paper examines the errors that the automatic speech recognition (ASR) system of Veed.io produces when transcribing utterances spoken in Jordanian Arabic into subtitles. It attempts to propose a new classification for the subtitles that are built based on artificial intelligence technology. Through a combination of qualitative and quantitative analyses, the study examines the types of errors and their impact on comprehension. The errors observed in the generated subtitles based on linguistic and phonetic analysis are categorised into three main types: deletions, substitutions, and insertions. Furthermore, the quantitative analysis measures the word error rate (WER) and shows that the WER percentage is 38.857% revealing that deletions are the most common type of error, followed by substitutions and insertions. The study recommends conducting further research on ASR systems for Arabic language dialects and advises subtitlers to be aware of the limitations of these systems when using them, ensuring that they edit and supervise them appropriately.Este artigo examina os erros que o sistema de reconhecimento automático de fala (ASR) do Veed.io produz ao transcrever declarações faladas em árabe jordaniano para legendas. Tenta propor uma nova classificação para as legendas construídas com base em tecnologia de inteligência artificial. Através de uma combinação de análises qualitativas e quantitativas, o estudo examina os tipos de erros e seu impacto na compreensão. Os erros observados nas legendas geradas com base na análise linguística e fonética são categorizados em três tipos principais: exclusões, substituições e inserções. Além disso, a análise quantitativa mede a taxa de erro de palavras (WER) e mostra que o percentual de WER é de 38,857%, revelando que as exclusões são o tipo de erro mais comum, seguidas pelas substituições e inserções. O estudo recomenda a realização de mais pesquisas sobre sistemas ASR para dialetos da língua árabe e aconselha os legendadores a estarem cientes das limitações desses sistemas ao usá-los, garantindo que os editem e supervisionem adequadamente.Universidade Federal de Minas Gerais2024-01-06info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionArtigo avaliado pelos paresapplication/pdfhttps://periodicos.ufmg.br/index.php/textolivre/article/view/4695210.1590/1983-3652.2024.46952Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e46952Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e46952Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e46952Texto Livre; v. 17 (2024): Texto Livre: Linguagem e Tecnologia; e469521983-3652reponame:Texto livreinstname:Universidade Federal de Minas Gerais (UFMG)instacron:UFMGenghttps://periodicos.ufmg.br/index.php/textolivre/article/view/46952/39657Copyright (c) 2024 Wala’ Mohammad Akasheh, Ahmad S. Haider, Bassam Al-Saideen, Yousef Saharihttps://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccessAkasheh, Wala’ MohammadHaider, Ahmad S.Al-Saideen, BassamSahari, Yousef2024-05-12T20:28:08Zoai:periodicos.ufmg.br:article/46952Revistahttp://www.periodicos.letras.ufmg.br/index.php/textolivrePUBhttps://periodicos.ufmg.br/index.php/textolivre/oairevistatextolivre@letras.ufmg.br1983-36521983-3652opendoar:2024-05-12T20:28:08Texto livre - Universidade Federal de Minas Gerais (UFMG)false |
dc.title.none.fl_str_mv |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic Legendas em árabe geradas por inteligência artificial: insights do sistema de reconhecimento automático de fala do árabe jordaniano da Veed.io |
title |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic |
spellingShingle |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic Akasheh, Wala’ Mohammad Subtitles Auto-generated subtitles Automatic Speech Recognition Linguistics Jordanian Arabic Legendas Legendas geradas automaticamente Reconhecimento Automático de Fala Linguística Árabe jordaniano |
title_short |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic |
title_full |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic |
title_fullStr |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic |
title_full_unstemmed |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic |
title_sort |
Artificial intelligence-generated Arabic subtitles: insights from Veed.io's automatic speech recognition system of Jordanian Arabic |
author |
Akasheh, Wala’ Mohammad |
author_facet |
Akasheh, Wala’ Mohammad Haider, Ahmad S. Al-Saideen, Bassam Sahari, Yousef |
author_role |
author |
author2 |
Haider, Ahmad S. Al-Saideen, Bassam Sahari, Yousef |
author2_role |
author author author |
dc.contributor.author.fl_str_mv |
Akasheh, Wala’ Mohammad Haider, Ahmad S. Al-Saideen, Bassam Sahari, Yousef |
dc.subject.por.fl_str_mv |
Subtitles Auto-generated subtitles Automatic Speech Recognition Linguistics Jordanian Arabic Legendas Legendas geradas automaticamente Reconhecimento Automático de Fala Linguística Árabe jordaniano |
topic |
Subtitles Auto-generated subtitles Automatic Speech Recognition Linguistics Jordanian Arabic Legendas Legendas geradas automaticamente Reconhecimento Automático de Fala Linguística Árabe jordaniano |
description |
This paper examines the errors that the automatic speech recognition (ASR) system of Veed.io produces when transcribing utterances spoken in Jordanian Arabic into subtitles. It attempts to propose a new classification for the subtitles that are built based on artificial intelligence technology. Through a combination of qualitative and quantitative analyses, the study examines the types of errors and their impact on comprehension. The errors observed in the generated subtitles based on linguistic and phonetic analysis are categorised into three main types: deletions, substitutions, and insertions. Furthermore, the quantitative analysis measures the word error rate (WER) and shows that the WER percentage is 38.857% revealing that deletions are the most common type of error, followed by substitutions and insertions. The study recommends conducting further research on ASR systems for Arabic language dialects and advises subtitlers to be aware of the limitations of these systems when using them, ensuring that they edit and supervise them appropriately. |
publishDate |
2024 |
dc.date.none.fl_str_mv |
2024-01-06 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Artigo avaliado pelos pares |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://periodicos.ufmg.br/index.php/textolivre/article/view/46952 10.1590/1983-3652.2024.46952 |
url |
https://periodicos.ufmg.br/index.php/textolivre/article/view/46952 |
identifier_str_mv |
10.1590/1983-3652.2024.46952 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
https://periodicos.ufmg.br/index.php/textolivre/article/view/46952/39657 |
dc.rights.driver.fl_str_mv |
Copyright (c) 2024 Wala’ Mohammad Akasheh, Ahmad S. Haider, Bassam Al-Saideen, Yousef Sahari https://creativecommons.org/licenses/by/4.0 info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Copyright (c) 2024 Wala’ Mohammad Akasheh, Ahmad S. Haider, Bassam Al-Saideen, Yousef Sahari https://creativecommons.org/licenses/by/4.0 |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal de Minas Gerais |
publisher.none.fl_str_mv |
Universidade Federal de Minas Gerais |
dc.source.none.fl_str_mv |
Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e46952 Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e46952 Texto Livre; Vol. 17 (2024): Texto Livre: Linguagem e Tecnologia; e46952 Texto Livre; v. 17 (2024): Texto Livre: Linguagem e Tecnologia; e46952 1983-3652 reponame:Texto livre instname:Universidade Federal de Minas Gerais (UFMG) instacron:UFMG |
instname_str |
Universidade Federal de Minas Gerais (UFMG) |
instacron_str |
UFMG |
institution |
UFMG |
reponame_str |
Texto livre |
collection |
Texto livre |
repository.name.fl_str_mv |
Texto livre - Universidade Federal de Minas Gerais (UFMG) |
repository.mail.fl_str_mv |
revistatextolivre@letras.ufmg.br |
_version_ |
1799711141135384576 |