Lightly supervised and unsupervised acoustic model training
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The guidelines caution against the common practice of trimming non-speech audio during automatic speech recognition model training.
“and caution against the common practice of trimming non-speech audio during automatic speech recognition (ASR) model training [56].”
From Disability-First AI Dataset Annotation: Co-designing Stuttered Speech Annotation Guidelines with People Who Stutter (2026) · cites this paper as [56]
Even when avoiding breaking a word or sentence, timestamps from machine transcription services could skip over subtle blocks or prolongations occurring just before or after a word or sentence, as ASR models typically neglect nonspeech content.
“Even when we avoided breaking a word or a sentence, the timestamps provided by machine transcription services could still skip over subtle blocks or prolongations occurring just before or after a word or sentence as ASR models typically neglect nonspeech content [56] (see Figure 2 for an example).”
From Disability-First AI Dataset Annotation: Co-designing Stuttered Speech Annotation Guidelines with People Who Stutter (2026) · cites this paper as [56]
Cited by
Sorted by citations
Self-supervised speech representation learning: A review
Mohamed, A., Lee, H.-Y., Borgholt, L., Havtorn, J. D., Edin, J., Igel, C., Kirchhoff, K., Li, S.-W., Livescu, K., Maaløe, L., et al. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3207050 · Cited by 372 · see citing works
“relies on a supervised teacher model to label a large volume of speech-only data, which is then used to augment the initial labeled data to train a student model [4], [5], [6], [61].”
MethodTang, X., Li, J., & Wu, S. (2026) · Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3790405 · Cited by 4 · see citing works
“…speech [58, 83], our guidelines draw on the lived experiences of PWS to high-light the importance of non-verbal signals (e.g. breathing, pauses) in communication, and caution against the common practice of trimming non-speech audio during automatic speech recognition (ASR) model training [56].”
- Authors
- Lamel, L., Gauvain, J.-L., & Adda, G.
- Affiliations
- Centre National de la Recherche Scientifique (FR)
- Venue
- Computer Speech & Language
- Publisher
- Elsevier BV
- Published
- 2002-01
- Volume
- 16
- Issue
- 1
- Pages
- 115-129
- ISSN
- 0885-2308 (ISSN-L)
- OpenAlex
- W1975113979
- Cited by
- 258
- References
- 0
- Citations by year
- 2026: 1 · 2025: 3 · 2024: 4 · 2023: 5 · 2022: 7 · 2021: 13 · 2020: 13 · 2019: 14 · 2018: 9 · 2017: 12 · 2016: 13 · 2015: 17 · 2014: 21 · 2013: 23 · 2012: 11
- FWCI
- 9.35
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Computer Speech & Language (publishedVersion)
- HAL (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Phonetics and Phonology Research
- Keywords
- acoustic model training, unsupervised learning, broadcast news, automatic transcription, closed captioning, language model training data, annotation cost reduction, speech recognition
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
