“Librispeech: an ASR corpus based on public domain audio books,”
The DOI matches its Crossref record; 22 of the 24 references it cites were found in a registry or an index.
Abstract
This paper introduces a new corpus of read English speech, suitable for training and evaluating speech recognition systems. The LibriSpeech corpus is derived from audiobooks that are part of the LibriVox project, and contains 1000 hours of speech sampled at 16 kHz. We have made the corpus freely available for download, along with separately prepared language-model training data and pre-built language models. We show that acoustic models trained on LibriSpeech give lower error rate on the Wall Street Journal (WSJ) test sets than models trained on WSJ itself. We are also releasing Kaldi scripts that make it easy to build these systems.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
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?
In the citing paper
“Often in ASR, to prevent equal weighing of short and long sentences, error rates are computed over paragraphs or longer segments of text143,144 rather than across individual sentences.”
From The speech neuroprosthesis (2024) · cites this paper as [143]
In the citing paper
“For instance, although the LibriSpeech (Panayotov et al., 2015) dataset contains a lot more data than VCTK (Yamagishi et al., 2019), Deep Voice 3 has significantly better quality on the VCTK dataset. One of the affecting factors could be the sampling rate which is 48 kHz for VCTK (Yamagishi et al., 2019), but just 16 kHZ for the LibriSpeech (Panayotov et al., 2015) dataset.”
From Audio deepfakes: A survey (2023) · cites this paper as Panayotov et al., 2015
In the citing paper
“Because challenge participants in the open condition were permitted to use models pre-trained using Librispeech, we removed all data collected from speakers appearing in the evaluation set when they also appear in Librispeech.”
From ASVspoof 5: crowdsourced speech data, deepfakes, and adversarial attacks at scale (2024) · cites this paper as [15]
In the citing paper
“Data for all speakers in the evaluation set, including both targets and non-targets, are removed if data corresponding to the same speakers also appears in the Librispeech database (Panayotov et al., 2015).”
From ASVspoof 5: Design, collection and validation of resources for spoofing, deepfake, and adversarial attack detection using crowdsourced speech (2026) · cites this paper as Panayotov et al., 2015
In the citing paper
“The LibriSpeech corpus [Panayotov et al. 2015] is a widely used benchmark in science and industry.”
From Measuring the Accuracy of Automatic Speech Recognition Solutions (2023) · cites this paper as Panayotov et al. 2015
5 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Binary Codes capable of currecting deletions, insertions, and reversals
Levenshtein, V. I. (1965) · Soviet physics. Doklady · Cited by 14,804
Identification of common molecular subsequences
Smith, T. F., & Waterman, M. S. (1981) · Journal of Molecular Biology · doi:10.1016/0022-2836(81)90087-5 · Cited by 9,940
Davis, S., & Mermelstein, P. (1980) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1980.1163420 · Cited by 5,465 · cited 2×
“For this first decoding pass we use a triphone model discriminatively trained with Boosted MMI [12], based on MFCC [13] features processed with frame-splicing over 7 frames, followed by LDA, followed by a global semi-tied covariance (STC) transform [14] .”
MethodKaldi Speech Recognition Toolkit
Povey, D. (2024) · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · doi:10.57702/jb3fvbn9 · Cited by 4,609
Self-citationSRILM - an extensible language modeling toolkit
Stolcke, A. (2002) · 7th International Conference on Spoken Language Processing (ICSLP 2002) · doi:10.21437/icslp.2002-303 · Cited by 4,326
Improved backing-off for M-gram language modeling
Kneser, R., & Ney, H. (2002) · 1995 International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp.1995.479394 · Cited by 1,473
The design for the wall street journal-based CSR corpus
Paul, D. B., & Baker, J. M. (1992) · Proceedings of the workshop on Speech and Natural Language - HLT '91 · doi:10.3115/1075527.1075614 · Cited by 1,099
An empirical study of smoothing techniques for language modeling
Chen, S. F., & Goodman, J. (1996) · Proceedings of the 34th annual meeting on Association for Computational Linguistics - · doi:10.3115/981863.981904 · Cited by 770
Witten, I. H., & Bell, T. C. (1991) · IEEE Transactions on Information Theory · doi:10.1109/18.87000 · Cited by 729
An Empirical Study of Smoothing Techniques for Language Modeling
Chen, S. F., & Goodman, J. (1996) · arXiv (Cornell University) · doi:10.48550/arxiv.cmp-lg/9606011 · Cited by 612
Research works printed but not linked: 2
- king, 2012, The Blizzard Challenge 2012, Proc Blizzard Workshop
- kneser, 1995, Improvedbacking-offform-gram language modeling, ICASSP, 1, 181
Cited by
Representation 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
“tions, and in reinforcement learning we measure how the auxiliary CPC loss speeds up learning of the agent. 3.1 Audio For audio, we use a 100-hour subset of the publicly available LibriSpeech dataset [30]. Although the dataset does not provide labels other than the raw text, we obtained force-aligned phone sequences 4…”
MethodInfluentialSpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
Park, D. S., Chan, W., Zhang, Y., Chiu, C.-C., Zoph, B., Cubuk, E. D., & Le, Q. V. (2019) · Interspeech 2019 · doi:10.21437/interspeech.2019-2680 · Cited by 3,558
Conformer: Convolution-augmented Transformer for Speech Recognition
Gulati, A., Qin, J., Chiu, C.-C., Parmar, N., Zhang, Y., Yu, J., Han, W., Wang, S., Zhang, Z., Wu, Y., et al. (2020) · Interspeech 2020 · doi:10.21437/interspeech.2020-3015 · Cited by 3,150
Conformer: Convolution-augmented Transformer for Speech Recognition
Gulati, A., Qin, J., Chiu, C.-C., Parmar, N., Zhang, Y., Yu, J., Han, W., Wang, S., Zhang, Z., Wu, Y., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.08100 · Cited by 3,150
“Table 2 compares the (WER) result of our model on LibriSpeech test-clean/test-other with a few state-of-the-art models include: ContextNet [10], Transformer transducer [7], and QuartzNet [9].”
MethodInfluentialHuBERT: 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
Metadata
- Authors
- Panayotov, V., Chen, G., Povey, D., & Khudanpur, S.
- Affiliations
- Johns Hopkins University (US)
- Venue
- 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Publisher
- IEEE
- Published
- 2015-04
- Pages
- 5206-5210
- ISSN
- 1520-6149 (ISSN-L)
- OpenAlex
- W1494198834
- Cited by
- 6,072
- References
- 24
- Citations by year
- 2026: 420 · 2025: 853 · 2024: 853 · 2023: 1022 · 2022: 770 · 2021: 934 · 2020: 600 · 2019: 348 · 2018: 156 · 2017: 80 · 2016: 25 · 2015: 11
- FWCI
- 51.10
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Acoustics Speech and Signal Processing (publishedVersion, public-domain)
- 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 › Artificial Intelligence › Speech and dialogue systems
- Keywords
- LibriSpeech, audiobooks, LibriVox, acoustic model training, language models, Wall Street Journal, Kaldi
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: “Librispeech: an ASR corpus based on public domain audio books,”. https://app.cochat.ai/papers/10.1109/icassp.2015.7178964
- The paper
- Panayotov, V., Chen, G., Povey, D., & Khudanpur, S. (2015). “Librispeech: an ASR corpus based on public domain audio books,”. 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). https://doi.org/10.1109/icassp.2015.7178964
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.