Papers › wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations

wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations

20 Jun 2020NeurIPS 2020 12arXiv:2006.11477archive 2025-07-28

Alexei Baevski, Henry Zhou, Abdel-rahman Mohamed, Michael Auli

We show for the first time that learning powerful representations from speech audio alone followed by fine-tuning on transcribed speech can outperform the best semi-supervised methods while being conceptually simpler. wav2vec 2.0 masks the speech input in the latent space and solves a contrastive task defined over a quantization of the latent representations which are jointly learned. Experiments using all labeled data of Librispeech achieve 1.8/3.3 WER on the clean/other test sets. When lowering the amount of labeled data to one hour, wav2vec 2.0 outperforms the previous state of the art on the 100 hour subset while using 100 times less labeled data. Using just ten minutes of labeled data and pre-training on 53k hours of unlabeled data still achieves 4.8/8.2 WER. This demonstrates the feasibility of speech recognition with limited amounts of labeled data.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2006.11477")

Code

Syntology Ran 2 of 9 code samples harvested from 3 repositories linked to this paper; 7 have no recorded run. Of those that ran: 2 ran · our draft was wrong.

By repository: community (archive-listed): 9 samples from 3 repositories, 2 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

25 repositories listed; official and paper-mentioned ones first.

pytorch/fairseq officialmentioned in paperpytorch report
AIdeaLab/wav2vec2_docker mentioned on GitHubpytorchMIT report
Arizona-Voice/Arizona-spotting mentioned on GitHubpytorchMIT report
BirgerMoell/tmh mentioned on GitHubpytorchMIT report
HarunoriKawano/Wav2vec2.0 mentioned on GitHubpytorchApache-2.0 report
JoungheeKim/Non-Attentive-Tacotron mentioned on GitHubpytorchApache-2.0 report
eastonYi/wav2vec mentioned on GitHubpytorch report
facebookresearch/brainmagick mentioned on GitHubpytorch report
gatech-eic/s3-router mentioned on GitHubpytorchMIT report
huggingface/transformers mentioned on GitHubpytorch report
huseinzol05/malaya-speech mentioned on GitHubtfMIT report
liutianlin0121/seislm mentioned on GitHubpytorch report
mailong25/vietnamese-speech-recognition mentioned on GitHubpytorch report
neonbjb/ocotillo mentioned on GitHubpytorchNOASSERTION report
nlp-en-es/wav2vec2-spanish mentioned on GitHubjax report
phanxuanphucnd/Arizona-asr mentioned on GitHubpytorchMIT report
phanxuanphucnd/wav2asr mentioned on GitHubMIT report
sh-lee-prml/hierspeechpp mentioned on GitHubpytorch report
shivangi-aneja/FaceTalk mentioned on GitHubpytorchNOASSERTION report
vasudevgupta7/gsoc-wav2vec2 mentioned on GitHubtfApache-2.0 report
pwc-1/Paper-9 mindspore report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

9 samples harvested; 2 ran; 0 honoured the contract we drafted; 7 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · our draft was wrong
7unverified

Licence: 2 of the 9 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 repositories linked to this paper, official or community; each sample names its own and says which. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

add_asr_eval_argument mailong25/self-supervised-speech-recognition/stt.py community (archive-listed) ran · our draft was wrong no licence file found · pointer only · 0f3c0e3f4461c02d · report
get_dataset_itr mailong25/self-supervised-speech-recognition/stt.py community (archive-listed) ran · our draft was wrong no licence file found · pointer only · 732cb84e2f814ba6 · report
apply_spec_augmentation vasudevgupta7/gsoc-wav2vec2/src/wav2vec2/spec_augment.py community (archive-listed) unverified Apache-2.0 (permissive) · fa1c3918c88bf8f5 · report
get_from_registry phanxuanphucnd/wav2asr/arizona/utils/misc_utils.py community (archive-listed) unverified MIT (permissive) · 90190d4490581c2d · report
ifnone phanxuanphucnd/wav2asr/arizona/utils/misc_utils.py community (archive-listed) unverified MIT (permissive) · ea6f42601b6cb44e · report
read_tfrecords vasudevgupta7/gsoc-wav2vec2/src/data_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · d39dd4b7bcc000b8 · report
replace vasudevgupta7/gsoc-wav2vec2/src/convert_torch_to_tf.py community (archive-listed) unverified Apache-2.0 (permissive) · e488217cd68c52ed · report
str2bool phanxuanphucnd/wav2asr/arizona/utils/misc_utils.py community (archive-listed) unverified MIT (permissive) · 9e0b1e5fd4f77016 · report
tf_multinomial_no_replacement vasudevgupta7/gsoc-wav2vec2/src/wav2vec2/spec_augment.py community (archive-listed) unverified Apache-2.0 (permissive) · d1e7bdb6b6c8d6cb · report

Tasks

QuantizationSelf-Supervised LearningSpeech RecognitionZero-Shot Audio Retrieval

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Recognition Libri-Light test-clean wav2vec 2.0 Large-10h-LV-60k Word Error Rate (WER) 2.5 #1 of 5 Archive leaderboard report
Speech Recognition Libri-Light test-other wav2vec 2.0 Large-10h-LV-60k Word Error Rate (WER) 5.0 #1 of 5 Archive leaderboard report
Speech Recognition LibriSpeech test-clean wav2vec 2.0 with Libri-Light Word Error Rate (WER) 1.8 #12 of 64 Archive leaderboard report
Speech Recognition LibriSpeech test-other wav2vec 2.0 with Libri-Light Word Error Rate (WER) 3.0 #6 of 53 Archive leaderboard report
Speech Recognition LibriSpeech test-other wav2vec 2.0 Word Error Rate (WER) 4.1 #17 of 53 Archive leaderboard report
Speech Recognition TIMIT wav2vec 2.0 Percentage error 8.3 #1 of 22 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGumbel SoftmaxLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections