Datasets › LibriSpeech

LibriSpeech

Introduced in Librispeech: An ASR corpus based on public domain audio books1 Jan 2015 archive 2025-07-28

The LibriSpeech corpus is a collection of approximately 1,000 hours of audiobooks that are a part of the LibriVox project. Most of the audiobooks come from the Project Gutenberg. The training data is split into 3 partitions of 100hr, 360hr, and 500hr sets while the dev and test data are split into the ’clean’ and ’other’ categories, respectively, depending upon how well or challenging Automatic Speech Recognition systems would perform against. Each of the dev and test sets is around 5hr in audio length. This corpus also provides the n-gram language models and the corresponding texts excerpted from the Project Gutenberg books, which contain 803M tokens and 977K unique words.

Source: State-of-the-art Speech Recognition using Multi-stream Self-attention with Dilated 1D Convolutions

Benchmarks archive 2025-07-28

All 4 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

Syntology 18 newer papers were judged to report on this dataset's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): Speech Recognition · LibriSpeech test-clean (18 judged, 0 placed), Speech Recognition · LibriSpeech test-other (8 judged, 0 placed), Voice Conversion · LibriSpeech test-clean (2 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

First row (archive order)PaperCode
Speech Recognition LibriSpeech test-clean United Med ASR Word Error Rate (WER) 0.985 High-precision medical speech recognition through... — 64 Compare
Speech Recognition LibriSpeech test-other SAMBA ASR Word Error Rate (WER) 2.48 Samba-ASR: State-Of-The-Art Speech Recognition... — 53 Compare
Voice Conversion LibriSpeech test-clean kNN-VC (prematched HiFiGAN) Character Error Rate (CER) 2.96 Voice Conversion With Just Nearest Neighbors bshall/knn-vc 1 Compare
Speech Recognition Librispeech (other) no rows — — 0 Compare

Papers archive 2025-07-28

papers with a benchmark row: 53 · with a code link: 39 · where Syntology ran a sample: 16 (15 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument) Syntology

Show: all papers with a benchmark rowonly where code ran (16 of 53 with a benchmark row: 15 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument)

Page 1 of 1: papers 1 to 53 of 53 with a leaderboard row on this dataset's benchmarks, newest first by the archive's date (ties by slug; undated papers last).

The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset, not that list; the archive's count for this dataset is 2,361. The Syntology column is from Syntology's graph, stated per sample; it is not part of any archive number. A line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified” (C is a part of N, never taken away from it); the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced. When the archive marks a repository official for the paper, the cell starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record.

PaperCodeResultsDateSamples run Syntology
Let SSMs be ConvNets: State-space Modeling with Optimal Tensor Contractions 0 1 22 Jan 2025 not harvested
Samba-ASR: State-Of-The-Art Speech Recognition Leveraging Structured State-Space Models 0 2 6 Jan 2025 not harvested
High-precision medical speech recognition through synthetic data and semantic correction: UNITED-MEDASR 0 1 24 Nov 2024 not harvested
CR-CTC: Consistency regularization on CTC for improved speech recognition 1 4 7 Oct 2024 not harvested
FAdam: Adam is a natural gradient optimizer using diagonal empirical Fisher information 1 2 21 May 2024 not harvested
Graph Convolutions Enrich the Self-Attention in Transformers! 1 2 7 Dec 2023 official (archive's flag): 6 ran · 22 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 2 honoured, 5 violated, 4 with no contract checked; 11 where Syntology's instrument failed) · 7 unverified (10 pointer-only for licence)
Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models 2 2 14 Nov 2023 official (archive's flag): 5 ran · 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 3 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified (7 pointer-only for licence)
Zipformer: A faster and better encoder for automatic speech recognition 1 2 17 Oct 2023 not harvested
Voice Conversion With Just Nearest Neighbors 1 1 30 May 2023 not harvested
Multi-Head State Space Model for Speech Recognition 0 1 21 May 2023 not harvested
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition 0 1 8 May 2023 not harvested
MT4SSL: Boosting Self-Supervised Speech Representation Learning by Integrating Multiple Targets 1 2 14 Nov 2022 not harvested
E-Branchformer: Branchformer with Enhanced merging for speech recognition 1 2 30 Sep 2022 not harvested
Squeezeformer: An Efficient Transformer for Automatic Speech Recognition 4 2 2 Jun 2022 official (archive's flag): 5 ran · 31 ran (of which 12 constructed an object rather than computing a result; 28 with no instrument failure: 3 honoured, 0 violated, 25 with no contract checked; 3 where Syntology's instrument failed) · 18 unverified
data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language 12 1 7 Feb 2022 community repositories only · 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified
WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing 9 2 26 Oct 2021 not harvested
W2v-BERT: Combining Contrastive Learning and Masked Language Modeling for Self-Supervised Speech Pre-Training 4 2 7 Aug 2021 not harvested
Amortized Neural Networks for Low-Latency Speech Recognition 0 1 3 Aug 2021 not harvested
Relaxed Attention: A Simple Method to Boost Performance of End-to-End Automatic Speech Recognition 1 1 2 Jul 2021 not harvested
HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units 11 2 14 Jun 2021 community repositories only · 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified
Librispeech Transducer Model with Internal Language Model Prior Correction 2 2 7 Apr 2021 not harvested
SpeechStew: Simply Mix All Available Speech Recognition Data to Train One Large Neural Network 0 4 5 Apr 2021 not harvested
Transformer-based ASR Incorporating Time-reduction Layer and Fine-tuning with Self-Knowledge Distillation 0 1 17 Mar 2021 not harvested
Improving RNN Transducer Based ASR with Auxiliary Tasks 1 2 5 Nov 2020 not harvested
Self-training and Pre-training are Complementary for Speech Recognition 3 3 22 Oct 2020 not harvested
Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition 1 2 20 Oct 2020 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified (2 pointer-only for licence)
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations 25 3 20 Jun 2020 community repositories only · 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified (2 pointer-only for licence)
ASAPP-ASR: Multistream CNN and Self-Attentive SRU for SOTA Speech Recognition 0 2 21 May 2020 not harvested
Faster, Simpler and More Accurate Hybrid ASR Systems Using Wordpieces 0 2 19 May 2020 not harvested
Improved Noisy Student Training for Automatic Speech Recognition 1 2 19 May 2020 not harvested
Iterative Pseudo-Labeling for Speech Recognition 1 2 19 May 2020 not harvested
Conformer: Convolution-augmented Transformer for Speech Recognition 25 6 16 May 2020 6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 3 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified (2 pointer-only for licence)
ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context 6 6 7 May 2020 not harvested
Semi-Supervised Speech Recognition via Local Prior Matching 1 3 24 Feb 2020 not harvested
End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures 1 4 19 Nov 2019 not harvested
Transformer-based Acoustic Modeling for Hybrid Speech Recognition 0 2 22 Oct 2019 not harvested
State-of-the-Art Speech Recognition Using Multi-Stream Self-Attention With Dilated 1D Convolutions 1 2 1 Oct 2019 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified (1 pointer-only for licence)
Espresso: A Fast End-to-end Neural Speech Recognition Toolkit 1 2 18 Sep 2019 official: no sample here; runs from other or unrecorded repositories · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified (1 pointer-only for licence)
A Comparative Study on Transformer vs RNN in Speech Applications 2 2 13 Sep 2019 not harvested
RWTH ASR Systems for LibriSpeech: Hybrid vs Attention -- w/o Data Augmentation 2 2 8 May 2019 not harvested
SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition 30 4 18 Apr 2019 14 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified
CRF-based Single-stage Acoustic Modeling with CTC Topology 1 2 16 Apr 2019 not harvested
Jasper: An End-to-End Convolutional Neural Acoustic Model 10 4 5 Apr 2019 not harvested
On the Choice of Modeling Unit for Sequence-to-Sequence Speech Recognition 3 1 5 Feb 2019 not harvested
Fully Convolutional Speech Recognition 0 2 17 Dec 2018 not harvested
The PyTorch-Kaldi Speech Recognition Toolkit 11 1 19 Nov 2018 official (archive's flag): 2 ran · 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified (6 pointer-only for licence)
Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces 16 2 25 May 2018 official (archive's flag): 6 ran · 13 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified
Improved training of end-to-end attention models for speech recognition 14 1 8 May 2018 12 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified
Neural Network Language Modeling with Letter-based Features and Importance Sampling 0 2 15 Apr 2018 not harvested
Letter-Based Speech Recognition with Gated ConvNets 2 1 22 Dec 2017 not harvested
Improving End-to-End Speech Recognition with Policy Learning 0 1 19 Dec 2017 not harvested
Deep Speech 2: End-to-End Speech Recognition in English and Mandarin 35 2 8 Dec 2015 community repositories only · 33 ran (of which 0 constructed an object rather than computing a result; 32 with no instrument failure: 0 honoured, 1 violated, 31 with no contract checked; 1 where Syntology's instrument failed) · 6 unverified
QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions 15 2 community repositories only · 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified (2 pointer-only for licence)

Dataset loaders archive 2025-07-28

4 loaders as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

License archive 2025-07-28

CC BY 4.0

Modalities archive 2025-07-28

Languages archive 2025-07-28

Variants archive 2025-07-28

  • Kazakh Speech Corpus 2 (KSC2)
  • LibriSpeech test other
  • LibriSpeech test clean
  • CommonVoice (clean)
  • LibriSpeech and External
  • librispeech_asr
  • Librispeech (other)
  • Librispeech (clean)
  • LibriSpeech ASR
  • LibriSpeechLibri-Light test-othertest-other
  • LibriSpeech
  • LibriSpeech test-other
  • LibriSpeech test-clean

13 variant names, as the archive lists them.

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