Papers › Improved training of end-to-end attention models for speech recognition

Improved training of end-to-end attention models for speech recognition

8 May 2018arXiv:1805.03294archive 2025-07-28

Albert Zeyer, Kazuki Irie, Ralf Schlüter, Hermann Ney

Sequence-to-sequence attention-based models on subword units allow simple open-vocabulary end-to-end speech recognition. In this work, we show that such models can achieve competitive results on the Switchboard 300h and LibriSpeech 1000h tasks. In particular, we report the state-of-the-art word error rates (WER) of 3.54% on the dev-clean and 3.82% on the test-clean evaluation subsets of LibriSpeech. We introduce a new pretraining scheme by starting with a high time reduction factor and lowering it during training, which is crucial both for convergence and final performance. In some experiments, we also use an auxiliary CTC loss function to help the convergence. In addition, we train long short-term memory (LSTM) language models on subword units. By shallow fusion, we report up to 27% relative improvements in WER over the attention baseline without a language model.

PaperPDFCodeCode 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="1805.03294")

Code

Syntology Ran 0 of 12 code samples harvested from 2 repositories linked to this paper; 12 have no recorded run.

By repository: community (archive-listed): 12 samples from 2 repositories, 0 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

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

rwth-i6/returnn mentioned in papertfNOASSERTION report
rwth-i6/returnn-experiments mentioned in paper report
bobchennan/espnet mentioned on GitHubpytorchApache-2.0 report
creatorscan/espnet mentioned on GitHubpytorchApache-2.0 report
danoneata/espnet mentioned on GitHubpytorchApache-2.0 report
dhanya-e/Google_Indic mentioned on GitHubpytorchApache-2.0 report
jumon/espnet-1 mentioned on GitHubpytorchApache-2.0 report
jzmo/espnet mentioned on GitHubpytorchApache-2.0 report
marynader6/espnetV.8 mentioned on GitHubpytorchApache-2.0 report
marynader6/transformerTTS mentioned on GitHubpytorchApache-2.0 report
pengchengguo/espnet mentioned on GitHubpytorchApache-2.0 report
pzelasko/espnet mentioned on GitHubpytorchApache-2.0 report
victor45664/espnet mentioned on GitHubpytorchApache-2.0 report
vinitunni/CoupledLoss-LAS-ESPNet mentioned on GitHubpytorchApache-2.0 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

12 samples harvested; 0 ran; 0 honoured the contract we drafted; 12 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.

12unverified

Licence: 0 of the 12 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 2 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_results_to_json creatorscan/espnet/espnet/mt/mt_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · bbe192baf1d74dd1 · report
count_tokens bobchennan/espnet/espnet/lm/lm_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · ff899e7de6121891 · report
count_tokens creatorscan/espnet/espnet/lm/lm_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · e2309fb95e3f748c · report
end_detect bobchennan/espnet/espnet/nets/e2e_asr_common.py community (archive-listed) unverified Apache-2.0 (permissive) · db10d7f5f75d9faa · report
get_vgg2l_odim bobchennan/espnet/espnet/nets/e2e_asr_common.py community (archive-listed) unverified Apache-2.0 (permissive) · 34e1e172156ca868 · report
label_smoothing_dist bobchennan/espnet/espnet/nets/e2e_asr_common.py community (archive-listed) unverified Apache-2.0 (permissive) · 0d9f822390e7d551 · report
load_dataset creatorscan/espnet/espnet/lm/lm_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 826840e1c3869a70 · report
make_batchset bobchennan/espnet/espnet/asr/asr_mix_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 6de087d747415428 · report
make_lexical_tree bobchennan/espnet/espnet/lm/lm_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 2956cc98fe367b81 · report
parse_hypothesis creatorscan/espnet/espnet/mt/mt_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 48cef9e2f067bc77 · report
read_tokens bobchennan/espnet/espnet/lm/lm_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · b9cca15d6b089537 · report
read_tokens creatorscan/espnet/espnet/lm/lm_utils.py community (archive-listed) unverified Apache-2.0 (permissive) · dc5fc4e9e58970cb · report

Tasks

Language ModelingLanguage ModellingSpeech Recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Recognition LibriSpeech test-clean Seq-to-seq attention Word Error Rate (WER) 3.82 #51 of 64 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

CTC Loss

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