Improving Mispronunciation Detection with Wav2vec2-based Momentum Pseudo-Labeling for Accentedness and Intelligibility Assessment Article Swipe
Yang Mu
,
Kevin Hirschi
,
Stephen D. Looney
,
Okim Kang
,
John H. L. Hansen
·
YOU?
·
· 2022
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2203.15937
YOU?
·
· 2022
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2203.15937
Current leading mispronunciation detection and diagnosis (MDD) systems achieve promising performance via end-to-end phoneme recognition. One challenge of such end-to-end solutions is the scarcity of human-annotated phonemes on natural L2 speech. In this work, we leverage unlabeled L2 speech via a pseudo-labeling (PL) procedure and extend the fine-tuning approach based on pre-trained self-supervised learning (SSL) models. Specifically, we use Wav2vec 2.0 as our SSL model, and fine-tune it using original labeled L2 speech samples plus the created pseudo-labeled L2 speech samples. Our pseudo labels are dynamic and are produced by an ensemble of the online model on-the-fly, which ensures that our model is robust to pseudo label noise. We show that fine-tuning with pseudo labels achieves a 5.35% phoneme error rate reduction and 2.48% MDD F1 score improvement over a labeled-samples-only fine-tuning baseline. The proposed PL method is also shown to outperform conventional offline PL methods. Compared to the state-of-the-art MDD systems, our MDD solution produces a more accurate and consistent phonetic error diagnosis. In addition, we conduct an open test on a separate UTD-4Accents dataset, where our system recognition outputs show a strong correlation with human perception, based on accentedness and intelligibility.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2203.15937
- https://arxiv.org/pdf/2203.15937
- OA Status
- green
- Cited By
- 1
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4226508289
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4226508289Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2203.15937Digital Object Identifier
- Title
-
Improving Mispronunciation Detection with Wav2vec2-based Momentum Pseudo-Labeling for Accentedness and Intelligibility AssessmentWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2022Year of publication
- Publication date
-
2022-03-29Full publication date if available
- Authors
-
Yang Mu, Kevin Hirschi, Stephen D. Looney, Okim Kang, John H. L. HansenList of authors in order
- Landing page
-
https://arxiv.org/abs/2203.15937Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2203.15937Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2203.15937Direct OA link when available
- Concepts
-
Speech recognition, Computer science, Intelligibility (philosophy), Leverage (statistics), Word error rate, Artificial intelligence, Philosophy, EpistemologyTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
1Total citation count in OpenAlex
- Citations by year (recent)
-
2024: 1Per-year citation counts (last 5 years)
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4226508289 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2203.15937 |
| ids.doi | https://doi.org/10.48550/arxiv.2203.15937 |
| ids.openalex | https://openalex.org/W4226508289 |
| fwci | |
| type | preprint |
| title | Improving Mispronunciation Detection with Wav2vec2-based Momentum Pseudo-Labeling for Accentedness and Intelligibility Assessment |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T10201 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9994999766349792 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1702 |
| topics[0].subfield.display_name | Artificial Intelligence |
| topics[0].display_name | Speech Recognition and Synthesis |
| topics[1].id | https://openalex.org/T10860 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9940999746322632 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1711 |
| topics[1].subfield.display_name | Signal Processing |
| topics[1].display_name | Speech and Audio Processing |
| topics[2].id | https://openalex.org/T12031 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9829999804496765 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1702 |
| topics[2].subfield.display_name | Artificial Intelligence |
| topics[2].display_name | Speech and dialogue systems |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C28490314 |
| concepts[0].level | 1 |
| concepts[0].score | 0.6783674955368042 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q189436 |
| concepts[0].display_name | Speech recognition |
| concepts[1].id | https://openalex.org/C41008148 |
| concepts[1].level | 0 |
| concepts[1].score | 0.6550923585891724 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[1].display_name | Computer science |
| concepts[2].id | https://openalex.org/C60048801 |
| concepts[2].level | 2 |
| concepts[2].score | 0.5662012100219727 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q1433889 |
| concepts[2].display_name | Intelligibility (philosophy) |
| concepts[3].id | https://openalex.org/C153083717 |
| concepts[3].level | 2 |
| concepts[3].score | 0.5442169308662415 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q6535263 |
| concepts[3].display_name | Leverage (statistics) |
| concepts[4].id | https://openalex.org/C40969351 |
| concepts[4].level | 2 |
| concepts[4].score | 0.43119141459465027 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q3516228 |
| concepts[4].display_name | Word error rate |
| concepts[5].id | https://openalex.org/C154945302 |
| concepts[5].level | 1 |
| concepts[5].score | 0.38576871156692505 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[5].display_name | Artificial intelligence |
| concepts[6].id | https://openalex.org/C138885662 |
| concepts[6].level | 0 |
| concepts[6].score | 0.0 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q5891 |
| concepts[6].display_name | Philosophy |
| concepts[7].id | https://openalex.org/C111472728 |
| concepts[7].level | 1 |
| concepts[7].score | 0.0 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q9471 |
| concepts[7].display_name | Epistemology |
| keywords[0].id | https://openalex.org/keywords/speech-recognition |
| keywords[0].score | 0.6783674955368042 |
| keywords[0].display_name | Speech recognition |
| keywords[1].id | https://openalex.org/keywords/computer-science |
| keywords[1].score | 0.6550923585891724 |
| keywords[1].display_name | Computer science |
| keywords[2].id | https://openalex.org/keywords/intelligibility |
| keywords[2].score | 0.5662012100219727 |
| keywords[2].display_name | Intelligibility (philosophy) |
| keywords[3].id | https://openalex.org/keywords/leverage |
| keywords[3].score | 0.5442169308662415 |
| keywords[3].display_name | Leverage (statistics) |
| keywords[4].id | https://openalex.org/keywords/word-error-rate |
| keywords[4].score | 0.43119141459465027 |
| keywords[4].display_name | Word error rate |
| keywords[5].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[5].score | 0.38576871156692505 |
| keywords[5].display_name | Artificial intelligence |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2203.15937 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | public-domain |
| locations[0].pdf_url | https://arxiv.org/pdf/2203.15937 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | https://openalex.org/licenses/public-domain |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2203.15937 |
| locations[1].id | doi:10.48550/arxiv.2203.15937 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2203.15937 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5100542450 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Yang Mu |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Yang, Mu |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5050069117 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-0838-3494 |
| authorships[1].author.display_name | Kevin Hirschi |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Hirschi, Kevin |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5042537011 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | Stephen D. Looney |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Looney, Stephen D. |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5019197893 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-7721-5283 |
| authorships[3].author.display_name | Okim Kang |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Kang, Okim |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5057910370 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-1382-9929 |
| authorships[4].author.display_name | John H. L. Hansen |
| authorships[4].author_position | last |
| authorships[4].raw_author_name | Hansen, John H. L. |
| authorships[4].is_corresponding | False |
| has_content.pdf | True |
| has_content.grobid_xml | True |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2203.15937 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2022-05-05T00:00:00 |
| display_name | Improving Mispronunciation Detection with Wav2vec2-based Momentum Pseudo-Labeling for Accentedness and Intelligibility Assessment |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T10201 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9994999766349792 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1702 |
| primary_topic.subfield.display_name | Artificial Intelligence |
| primary_topic.display_name | Speech Recognition and Synthesis |
| related_works | https://openalex.org/W2127461790, https://openalex.org/W2069324367, https://openalex.org/W2309273277, https://openalex.org/W1769849273, https://openalex.org/W2061937230, https://openalex.org/W1574295218, https://openalex.org/W113247760, https://openalex.org/W2547793174, https://openalex.org/W2132885390, https://openalex.org/W2070212102 |
| cited_by_count | 1 |
| counts_by_year[0].year | 2024 |
| counts_by_year[0].cited_by_count | 1 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2203.15937 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | public-domain |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2203.15937 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | https://openalex.org/licenses/public-domain |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2203.15937 |
| primary_location.id | pmh:oai:arXiv.org:2203.15937 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | public-domain |
| primary_location.pdf_url | https://arxiv.org/pdf/2203.15937 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | https://openalex.org/licenses/public-domain |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2203.15937 |
| publication_date | 2022-03-29 |
| publication_year | 2022 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 40, 116, 129, 156, 172, 182 |
| abstract_inverted_index.F1 | 125 |
| abstract_inverted_index.In | 31, 164 |
| abstract_inverted_index.L2 | 29, 37, 71, 78 |
| abstract_inverted_index.PL | 135, 144 |
| abstract_inverted_index.We | 108 |
| abstract_inverted_index.an | 90, 168 |
| abstract_inverted_index.as | 61 |
| abstract_inverted_index.by | 89 |
| abstract_inverted_index.is | 21, 102, 137 |
| abstract_inverted_index.it | 67 |
| abstract_inverted_index.of | 17, 24, 92 |
| abstract_inverted_index.on | 27, 50, 171, 189 |
| abstract_inverted_index.to | 104, 140, 147 |
| abstract_inverted_index.we | 34, 57, 166 |
| abstract_inverted_index.2.0 | 60 |
| abstract_inverted_index.MDD | 124, 150, 153 |
| abstract_inverted_index.One | 15 |
| abstract_inverted_index.Our | 81 |
| abstract_inverted_index.SSL | 63 |
| abstract_inverted_index.The | 133 |
| abstract_inverted_index.and | 4, 44, 65, 86, 122, 159, 191 |
| abstract_inverted_index.are | 84, 87 |
| abstract_inverted_index.our | 62, 100, 152, 177 |
| abstract_inverted_index.the | 22, 46, 75, 93, 148 |
| abstract_inverted_index.use | 58 |
| abstract_inverted_index.via | 11, 39 |
| abstract_inverted_index.(PL) | 42 |
| abstract_inverted_index.also | 138 |
| abstract_inverted_index.more | 157 |
| abstract_inverted_index.open | 169 |
| abstract_inverted_index.over | 128 |
| abstract_inverted_index.plus | 74 |
| abstract_inverted_index.rate | 120 |
| abstract_inverted_index.show | 109, 181 |
| abstract_inverted_index.such | 18 |
| abstract_inverted_index.test | 170 |
| abstract_inverted_index.that | 99, 110 |
| abstract_inverted_index.this | 32 |
| abstract_inverted_index.with | 112, 185 |
| abstract_inverted_index.(MDD) | 6 |
| abstract_inverted_index.(SSL) | 54 |
| abstract_inverted_index.2.48% | 123 |
| abstract_inverted_index.5.35% | 117 |
| abstract_inverted_index.based | 49, 188 |
| abstract_inverted_index.error | 119, 162 |
| abstract_inverted_index.human | 186 |
| abstract_inverted_index.label | 106 |
| abstract_inverted_index.model | 95, 101 |
| abstract_inverted_index.score | 126 |
| abstract_inverted_index.shown | 139 |
| abstract_inverted_index.using | 68 |
| abstract_inverted_index.where | 176 |
| abstract_inverted_index.which | 97 |
| abstract_inverted_index.work, | 33 |
| abstract_inverted_index.extend | 45 |
| abstract_inverted_index.labels | 83, 114 |
| abstract_inverted_index.method | 136 |
| abstract_inverted_index.model, | 64 |
| abstract_inverted_index.noise. | 107 |
| abstract_inverted_index.online | 94 |
| abstract_inverted_index.pseudo | 82, 105, 113 |
| abstract_inverted_index.robust | 103 |
| abstract_inverted_index.speech | 38, 72, 79 |
| abstract_inverted_index.strong | 183 |
| abstract_inverted_index.system | 178 |
| abstract_inverted_index.Current | 0 |
| abstract_inverted_index.Wav2vec | 59 |
| abstract_inverted_index.achieve | 8 |
| abstract_inverted_index.conduct | 167 |
| abstract_inverted_index.created | 76 |
| abstract_inverted_index.dynamic | 85 |
| abstract_inverted_index.ensures | 98 |
| abstract_inverted_index.labeled | 70 |
| abstract_inverted_index.leading | 1 |
| abstract_inverted_index.models. | 55 |
| abstract_inverted_index.natural | 28 |
| abstract_inverted_index.offline | 143 |
| abstract_inverted_index.outputs | 180 |
| abstract_inverted_index.phoneme | 13, 118 |
| abstract_inverted_index.samples | 73 |
| abstract_inverted_index.speech. | 30 |
| abstract_inverted_index.systems | 7 |
| abstract_inverted_index.Compared | 146 |
| abstract_inverted_index.accurate | 158 |
| abstract_inverted_index.achieves | 115 |
| abstract_inverted_index.approach | 48 |
| abstract_inverted_index.dataset, | 175 |
| abstract_inverted_index.ensemble | 91 |
| abstract_inverted_index.learning | 53 |
| abstract_inverted_index.leverage | 35 |
| abstract_inverted_index.methods. | 145 |
| abstract_inverted_index.original | 69 |
| abstract_inverted_index.phonemes | 26 |
| abstract_inverted_index.phonetic | 161 |
| abstract_inverted_index.produced | 88 |
| abstract_inverted_index.produces | 155 |
| abstract_inverted_index.proposed | 134 |
| abstract_inverted_index.samples. | 80 |
| abstract_inverted_index.scarcity | 23 |
| abstract_inverted_index.separate | 173 |
| abstract_inverted_index.solution | 154 |
| abstract_inverted_index.systems, | 151 |
| abstract_inverted_index.addition, | 165 |
| abstract_inverted_index.baseline. | 132 |
| abstract_inverted_index.challenge | 16 |
| abstract_inverted_index.detection | 3 |
| abstract_inverted_index.diagnosis | 5 |
| abstract_inverted_index.fine-tune | 66 |
| abstract_inverted_index.procedure | 43 |
| abstract_inverted_index.promising | 9 |
| abstract_inverted_index.reduction | 121 |
| abstract_inverted_index.solutions | 20 |
| abstract_inverted_index.unlabeled | 36 |
| abstract_inverted_index.consistent | 160 |
| abstract_inverted_index.diagnosis. | 163 |
| abstract_inverted_index.end-to-end | 12, 19 |
| abstract_inverted_index.outperform | 141 |
| abstract_inverted_index.correlation | 184 |
| abstract_inverted_index.fine-tuning | 47, 111, 131 |
| abstract_inverted_index.improvement | 127 |
| abstract_inverted_index.on-the-fly, | 96 |
| abstract_inverted_index.perception, | 187 |
| abstract_inverted_index.performance | 10 |
| abstract_inverted_index.pre-trained | 51 |
| abstract_inverted_index.recognition | 179 |
| abstract_inverted_index.UTD-4Accents | 174 |
| abstract_inverted_index.accentedness | 190 |
| abstract_inverted_index.conventional | 142 |
| abstract_inverted_index.recognition. | 14 |
| abstract_inverted_index.Specifically, | 56 |
| abstract_inverted_index.pseudo-labeled | 77 |
| abstract_inverted_index.human-annotated | 25 |
| abstract_inverted_index.pseudo-labeling | 41 |
| abstract_inverted_index.self-supervised | 52 |
| abstract_inverted_index.intelligibility. | 192 |
| abstract_inverted_index.mispronunciation | 2 |
| abstract_inverted_index.state-of-the-art | 149 |
| abstract_inverted_index.labeled-samples-only | 130 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 5 |
| sustainable_development_goals[0].id | https://metadata.un.org/sdg/4 |
| sustainable_development_goals[0].score | 0.5 |
| sustainable_development_goals[0].display_name | Quality Education |
| citation_normalized_percentile |