Papers › End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures

End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures

19 Nov 2019arXiv:1911.08460archive 2025-07-28

Gabriel Synnaeve, Qiantong Xu, Jacob Kahn, Tatiana Likhomanenko, Edouard Grave, Vineel Pratap, Anuroop Sriram, Vitaliy Liptchinsky, Ronan Collobert

We study pseudo-labeling for the semi-supervised training of ResNet, Time-Depth Separable ConvNets, and Transformers for speech recognition, with either CTC or Seq2Seq loss functions. We perform experiments on the standard LibriSpeech dataset, and leverage additional unlabeled data from LibriVox through pseudo-labeling. We show that while Transformer-based acoustic models have superior performance with the supervised dataset alone, semi-supervision improves all models across architectures and loss functions and bridges much of the performance gaps between them. In doing so, we reach a new state-of-the-art for end-to-end acoustic models decoded with an external language model in the standard supervised learning setting, and a new absolute state-of-the-art with semi-supervised training. Finally, we study the effect of leveraging different amounts of unlabeled audio, propose several ways of evaluating the characteristics of unlabeled audio which improve acoustic modeling, and show that acoustic models trained with more audio rely less on external language models.

PaperPDFCode

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

Code

facebookresearch/wav2letter mentioned on GitHub 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

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Language ModelingLanguage ModellingSpeech Recognitionspeech-recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Recognition LibriSpeech test-clean Conv + Transformer AM + Pseudo-Labeling (ConvLM with Transformer Rescoring) Word Error Rate (WER) 2.03 #27 of 64 Archive leaderboard report
Speech Recognition LibriSpeech test-clean Conv + Transformer AM (ConvLM with Transformer Rescoring) (LS only) Word Error Rate (WER) 2.31 #37 of 64 Archive leaderboard report
Speech Recognition LibriSpeech test-other Conv + Transformer AM (ConvLM with Transformer Rescoring) Word Error Rate (WER) 4.11 #19 of 53 Archive leaderboard report
Speech Recognition LibriSpeech test-other Conv + Transformer AM (ConvLM with Transformer Rescoring) (LS only) Word Error Rate (WER) 5.18 #32 of 53 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

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionGlobal Average PoolingKaiming InitializationLSTMMax PoolingReLUResidual BlockResidual ConnectionSeq2SeqSigmoid ActivationTanh Activation

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