The DOI matches its Crossref record; 67 of the 69 references it cites were found in a registry or an index.
Word Segmentation on Discovered Phone Units With Dynamic Programming and Self-Supervised Scoring
Open accessAbstract
Recent work on unsupervised speech segmentation has used self-supervised models with phone and word segmentation modules that are trained jointly. This paper instead revisits an older approach to word segmentation: bottom-up phone-like unit discovery is performed first, and symbolic word segmentation is then performed on top of the discovered units (without influencing the lower level). To do this, I propose a new unit discovery model, a new symbolic word segmentation model, and then chain the two models to segment speech. Both models use dynamic programming to minimize segment costs from a self-supervised network with an additional duration penalty that encourages longer units. Concretely, for acoustic unit discovery, duration-penalized dynamic programming (DPDP) is used with a contrastive predictive coding model as the scoring network. For word segmentation, DPDP is applied with an autoencoding recurrent neural as the scoring network. The two models are chained in order to segment speech. This approach gives comparable word segmentation results to state-of-the-art joint self-supervised segmentation models on an English benchmark. On French, Mandarin, German and Wolof data, it outperforms previous systems on the ZeroSpeech benchmarks. Analysis shows that the chained DPDP system segments shorter filler words well, but longer words might require some external top-down signal.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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 cited source says what the claim attributes to it.
In the citing paper
“Interestingly, the discrete representations enable the unsupervised discovery of acoustic units where phonemes are automatically mapped to a small set of discrete representations, enabling phoneme discovery and segmentation [54,55,56,57].”
From Towards Learning Discrete Representations via Self-Supervision for Wearables-Based Human Activity Recognition (2024) · cites this paper as [55]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“The AE-RNN is trained with Adam optimization [54] using a learning rate of 1 · 10−3 for 1500 steps on the full tokenized utterances.”
MethodRepresentation Learning with Contrastive Predictive Coding
Oord, A. V. D., Li, Y., & Vinyals, O. (2018) · arXiv · doi:10.48550/arxiv.1807.03748 · Cited by 6,242 · cited 2×
“Standard CPC speech models learn continuous features by trying to classify future observations (at different time-steps) from among a set of negative examples [13].”
BackgroundHuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units
Hsu, W.-N., Bolte, B., Tsai, Y.-H. H., Lakhotia, K., Salakhutdinov, R., & Mohamed, A. (2021) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2021.3122291 · Cited by 2,983
Libri-Light: A Benchmark for ASR with Limited or No Supervision
Kahn, J., Riviere, M., Zheng, W., Kharitonov, E., Xu, Q., Mazare, P. E., Karadayi, J., Liptchinsky, V., Collobert, R., Fuegen, C., et al. (2020) · ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp40776.2020.9052942 · Cited by 458
A Bayesian framework for word segmentation: Exploring the effects of context
Goldwater, S., Griffiths, T. L., & Johnson, M. (2009) · Cognition · doi:10.1016/j.cognition.2009.03.008 · Cited by 431 · cited 3×
“(phoneme or phone sequences) [2], [3], [4].”
BackgroundMethodPitt, M. A., Johnson, K., Hume, E., Kiesling, S., & Raymond, W. (2005) · Speech Communication · doi:10.1016/j.specom.2004.09.001 · Cited by 390 · cited 11×
“For evaluation on English, testing is then done on the 5-hour test set from Buckeye.”
InfluentialAn Unsupervised Autoregressive Model for Speech Representation Learning
Chung, Y.-A., Hsu, W.-N., Tang, H., & Glass, J. (2019) · Interspeech 2019 · doi:10.21437/interspeech.2019-1473 · Cited by 357
An Efficient, Probabilistically Sound Algorithm for Segmentation and Word Discovery
Brent, M. R. (1999) · Machine Learning · doi:10.1023/a:1007541817488 · Cited by 351 · cited 2×
“I specifically consider results on the Brent corpus [48], a standard benchmark for symbolic word segmentation.”
BackgroundUnsupervised Pattern Discovery in Speech
Park, A. S., & Glass, J. R. (2008) · IEEE Transactions on Audio, Speech, and Language Processing · doi:10.1109/tasl.2007.909282 · Cited by 351 · cited 1×
“One approach would be to incorporate top-down information from a sparse term discovery system tailored towards discovering recurring but longer words [55], [56], [59], which could be used to bootstrap segmentation.”
BackgroundBayesian Nonparametric Hidden Semi-Markov ModelsPreprint
Johnson, M. J., & Willsky, A. S. (2012) · arXiv (Cornell University) · doi:10.48550/arxiv.1203.1365 · Cited by 335 · cited 1×
“To address this, probabilistic models that explicitly model per-unit durations could be integrated or adapted to the DPDP approaches proposed here [46], [67].”
Method
Research works printed but not linked: 4
- 10.1109/asru.2017.8269008 · doi:10.1109/asru.2017.8269008
- Baevski, 2020, vq-wav2vec: Self-supervised learning of discrete speech representations
- Nguyen, 2020, The Zero Resource Speech Benchmark 2021: Metrics and baselines for unsupervised spoken language modeling
- 10.18653/v1/2021.insights-1.11 · DOI
Cited by
What Do Self-Supervised Speech Models Know About Words?
Pasad, A., Chien, C.-M., Settle, S., & Livescu, K. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00656 · Cited by 55
“However, this line of work mostly relies on a complex algorithm on top of vanilla S3M representations, such as audio-visual contrastive training [50], dynamic programming [51] or a two-stage prediction process with pseudo labeling [52], and usually only focuses on a specific S3M.”
MethodWord Discovery in Visually Grounded, Self-Supervised Speech Models
Peng, P., & Harwath, D. F. (2022) · Interspeech (USB) · doi:10.21437/interspeech.2022-10652 · Cited by 41
Qi, H., Zhou, Z., Manu, P., & Li, N. (2025) · Safety Science · doi:10.1016/j.ssci.2025.106814 · Cited by 35
Self-Supervised Language Learning From Raw Audio: Lessons From the Zero Resource Speech Challenge
Dunbar, E., Hamilakis, N., & Emmanuel Dupoux, A. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3206084 · Cited by 31
“TDE-15 (Task 2) BAS2 2015 [32] JHU Match-first Kam15a 2015 [33] Stellenbosch Match-first Lyz15a-c 2015 [34] JHU Match-first Ras15a-c 2015 [35] Aalto Match-first TDE-17 (Task 2) BAS2 2017 [32] JHU Match-first Kam22 2022 [36] Stellenbosch Segment-first Alg22 2022 [37] CoML Segment-first Kam17 2017 [38] Stellenbosch Segm…”
DP-Parse: Finding Word Boundaries from Raw Speech with an Instance Lexicon
Algayres, R., Ricoul, T., Karadayi, J., Laurençon, H., Zaiem, S., Mohamed, A., Sagot, B., & Emmanuel Dupoux, A. (2022) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00505 · Cited by 17
“A recent unpublished paper (Kamper, 2022) came to our attention based on the non-lexical principle and showed similar or slightly better results than ours on a subset of the ZR17 language.”
Result
Metadata
- Authors
- Kamper, H.
- Affiliations
- Stellenbosch University (ZA)
- Corresponding author
- Kamper, Herman
- ORCID
- Kamper, Herman
- Venue
- IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Published
- 2023
- Volume
- 31
- Pages
- 684-694
- ISSN
- 2329-9290 (ISSN-L)
- 2329-9304
- OpenAlex
- W4313182775
- Cited by
- 24
- References
- 69
- Citations by year
- 2026: 2 · 2025: 5 · 2024: 5 · 2023: 7
- FWCI
- 1.81
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE/ACM Transactions on Audio Speech and Language Processing (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Research Foundation of South Africa (120409)
- 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
- word segmentation, self-supervised learning, dynamic programming, contrastive predictive coding, autoencoder, filler words
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Word Segmentation on Discovered Phone Units With Dynamic Programming and Self-Supervised Scoring. https://app.cochat.ai/papers/10.1109/taslp.2022.3229264
- The paper
- Kamper, H. (2023). Word Segmentation on Discovered Phone Units With Dynamic Programming and Self-Supervised Scoring. IEEE/ACM Transactions on Audio, Speech, and Language Processing. https://doi.org/10.1109/taslp.2022.3229264
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.