Self-Supervised Speech Representation Learning: A Review
Open accessThe DOI matches its Crossref record; 358 of the 359 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
Although supervised deep learning has revolutionized speech and audio processing, it has necessitated the building of specialist models for individual tasks and application scenarios. It is likewise difficult to apply this to dialects and languages for which only limited labeled data is available. Self-supervised representation learning methods promise a single universal model that would benefit a wide variety of tasks and domains. Such methods have shown success in natural language processing and computer vision domains, achieving new levels of performance while reducing the number of labels required for many downstream scenarios. Speech representation learning is experiencing similar progress in three main categories: generative, contrastive, and predictive methods. Other approaches rely on multi-modal data for pre-training, mixing text or visual data streams with speech. Although self-supervised speech representation is still a nascent research area, it is closely related to acoustic word embedding and learning with zero lexical resources, both of which have seen active research for many years. This review presents approaches for self-supervised speech representation learning and their connection to other research areas. Since many current methods focus solely on automatic speech recognition as a downstream task, we review recent efforts on benchmarking learned representations to extend the application beyond speech recognition.
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
“Many popular self-supervised learning (SSL) models (Mohamed et al., 2022, Table IV) are trained using the Librispeech database”
From ASVspoof 5: Design, collection and validation of resources for spoofing, deepfake, and adversarial attack detection using crowdsourced speech (2026) · cites this paper as Mohamed et al., 2022
In the citing paper
“A previous study [34] already proposed a method to adapt SimCLR to the HAR domain, finding that random 3D rotation was an effective transformation for accelerometer data.”
From Comparing self-supervised learning techniques for wearable human activity recognition (2025) · cites this paper as [34]
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
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173 · cited 4×
“c) Masked Reconstruction: Masked reconstruction is largely inspired by the masked language model (MLM) task from BERT [54].”
BackgroundMethodInfluential(2014) · Cited by 33,419
PyTorch: An Imperative Style, High-Performance Deep Learning Library
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al. (2019) · arXiv · doi:10.48550/arxiv.1912.01703 · Cited by 31,473 · cited 1×
“improve continuously via the EMA teacher network; however, student-teacher approaches inflate the existing computational challenges of very large models and may necessitate the use of methods that decrease instantaneous memory utilization such as mixed precision training, model parallelism and model sharding [142].”
Girshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105
Reducing the dimensionality of data with neural networks
Hinton, G. E., & Salakhutdinov, R. R. (2006) · Science · doi:10.1126/science.1127647 · Cited by 20,928
Auto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019) · arXiv · doi:10.48550/arxiv.1907.11692 · Cited by 16,515
Learning the parts of objects by non-negative matrix factorization
Lee, D. D., & Seung, H. S. (1999) · Nature · doi:10.1038/44565 · Cited by 13,868 · cited 1×
“niques include k-means clustering [11], mixture models [12], autoencoders [13], and non-negative matrix factorization [14].”
Momentum Contrast for Unsupervised Visual Representation Learning
He, K., Fan, H., Wu, Y., Xie, S., & Girshick, R. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00975 · Cited by 13,347 · cited 1×
“This approach has become especially popular in the CV context [57], [58], [59], [60].”
Background
Research works printed but not linked: 18
- 10.1109/massp.1984.1162229 · doi:10.1109/massp.1984.1162229
- Radford, 2019, Language models are unsupervised multitask learners, 1, 9
- 10.5555/3524938.3525087 · doi:10.5555/3524938.3525087
- Garofolo, 1993, TIMIT acoustic-phonetic continuous speech corpus LDC93S1
- Yang, 2019, Generalized Autoregressive Pretraining for Language Understanding, 32
- 10.1109/icassp40776.2020.9053176 · DOI
- Baevski, 2020, vq-wav2vec: Self-supervised learning of discrete speech representations
- Baevski, 2020, wav2vec 2.0: A framework for self-supervised learning of speech representations, 33, 12449
- 10.1007/978-3-030-01264-9_9 · DOI · Paper page
- 10.5555/3495724.3497510 · doi:10.5555/3495724.3497510
- 10.1109/icassp.2018.8461761 · DOI
- 10.1109/asru.2017.8269008 · doi:10.1109/asru.2017.8269008
- Maas, 2012, Word-level acoustic modeling with convolutional vector regression
- Rodrguez-Fuentes, 2014, GTTS-EHU systems for QUESST at MediaEval 2014
- 10.1109/icassp40776.2020.9054548 · DOI · Paper page
- Pu, 2021, Scaling effect of self-supervised models, 1084
- Goodfellow, 2014, Generative adversarial nets, 2672
- Arjovsky, 2017, GAN wasserstein, 214
Cited by
Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects
Zhang, K., Wen, Q., Zhang, C., Cai, R., Jin, M., Liu, Y., Zhang, J. Y., Liang, Y., Pang, G., Song, D., et al. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3387317 · Cited by 295
“Using the learning paradigm as a taxonomy is arguably the most popular among the existing SSL surveys, including [20], [22], [23], [24], [25], [26], [27].”
MethodTen deep learning techniques to address small data problems with remote sensing
Safonova, A., Ghazaryan, G., Stiller, S., Main-Knorn, M., Nendel, C., & Ryo, M. (2023) · International Journal of Applied Earth Observation and Geoinformation · doi:10.1016/j.jag.2023.103569 · Cited by 196
“, 2022), physics (Ma and Liu, 2020), speech representation (Mohamed et al., 2022), RS (S.”
BackgroundComparative Layer-Wise Analysis of Self-Supervised Speech Models
Pasad, A., Shi, B., & Livescu, K. (2023) · ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp49357.2023.10096149 · Cited by 105
“Self-supervised models have become a nearly ubiquitous approach for learning speech representations and improving performance on downstream tasks [1–5], but our understanding of their properties and strategies for their use is still limited.”
BackgroundSelf-citationGalić, I., Habijan, M., Leventić, H., & Romić, K. (2023) · Electronics · doi:10.3390/electronics12214411 · Cited by 150
Automated Medical Coding on MIMIC-III and MIMIC-IV: A Critical Review and Replicability Study
Edin, J., Junge, A., Havtorn, J. D., Borgholt, L., Maistro, M., Ruotsalo, T., & Maaløe, L. (2023) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3539618.3591918 · Cited by 137
“Finally, while PLM-ICD outperforms the other models in this paper, the improvements are limited compared to the effect of pre-training in other domains [1, 6, 9, 23, 28].”
Self-citation
Metadata
- Authors
- Mohamed, A., Lee, H.-Y., Borgholt, L., Havtorn, J. D., Edin, J., Igel, C., Kirchhoff, K., Li, S.-W., Livescu, K., Maaløe, L., Sainath, T. N., & Watanabe, S.
- Affiliations
- Meta (United States) (US)
- National Taiwan University (TW)
- University of Copenhagen (DK)
- Technical University of Denmark (DK)
- Amazon (United States) (US)
- Toyota Technological Institute at Chicago (US)
- Google (United States) (US)
- Carnegie Mellon University (US)
- ORCID
- Venue
- IEEE Journal of Selected Topics in Signal Processing
- Published
- 2022-10
- Volume
- 16
- Issue
- 6
- Pages
- 1179-1210
- ISSN
- 1932-4553 (ISSN-L)
- 1941-0484
- OpenAlex
- W4281492411
- Semantic Scholar
- f76a5b176f3435214eb87dd105f730f0b53672c3
- Cited by
- 372
- References
- 345
- Citations by year
- 2026: 66 · 2025: 91 · 2024: 110 · 2023: 94 · 2022: 11
- FWCI
- 5.35
- Type
- journal article
- Language
- en
- Open access
- Licence
- Locations
- IEEE Journal of Selected Topics in Signal Processing (publishedVersion)
- Research at the University of Copenhagen (University of Copenhagen) (publishedVersion, other-oa)
- arXiv (Cornell University) (submittedVersion)
- Technical University of Denmark, DTU Orbit (Technical University of Denmark, DTU) (submittedVersion, other-oa)
- arXiv (Cornell University) (cc-by)
- Funders
- Innovation Fund Denmark through Industrial Ph.D. Programme (0153-00167B)
- Innovation Fund Denmark thorugh Industrial Ph.D. Programme (8053-00184B)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- self-supervised learning, speech representation learning, generative models, contrastive learning, prediction model, multimodal pre-training, acoustic word embeddings, low-resource languages, 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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Self-Supervised Speech Representation Learning: A Review. https://app.cochat.ai/papers/10.1109/jstsp.2022.3207050
- The paper
- Mohamed, A., Lee, H.-Y., Borgholt, L., Havtorn, J. D., Edin, J., Igel, C., Kirchhoff, K., Li, S.-W., Livescu, K., Maaløe, L., Sainath, T. N., & Watanabe, S. (2022). Self-Supervised Speech Representation Learning: A Review. IEEE Journal of Selected Topics in Signal Processing. https://doi.org/10.1109/jstsp.2022.3207050
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.