Papers › Deep Speech 2: End-to-End Speech Recognition in English and Mandarin

Deep Speech 2: End-to-End Speech Recognition in English and Mandarin

8 Dec 2015arXiv:1512.02595archive 2025-07-28

Dario Amodei, Rishita Anubhai, Eric Battenberg, Carl Case, Jared Casper, Bryan Catanzaro, Jingdong Chen, Mike Chrzanowski, Adam Coates, Greg Diamos, Erich Elsen, Jesse Engel, Linxi Fan, Christopher Fougner, Tony Han, Awni Hannun, Billy Jun, Patrick LeGresley, Libby Lin, Sharan Narang, Andrew Ng, Sherjil Ozair, Ryan Prenger, Jonathan Raiman, Sanjeev Satheesh, David Seetapun, Shubho Sengupta, Yi Wang, Zhiqian Wang, Chong Wang, Bo Xiao, Dani Yogatama, Jun Zhan, Zhenyao Zhu

We show that an end-to-end deep learning approach can be used to recognize either English or Mandarin Chinese speech--two vastly different languages. Because it replaces entire pipelines of hand-engineered components with neural networks, end-to-end learning allows us to handle a diverse variety of speech including noisy environments, accents and different languages. Key to our approach is our application of HPC techniques, resulting in a 7x speedup over our previous system. Because of this efficiency, experiments that previously took weeks now run in days. This enables us to iterate more quickly to identify superior architectures and algorithms. As a result, in several cases, our system is competitive with the transcription of human workers when benchmarked on standard datasets. Finally, using a technique called Batch Dispatch with GPUs in the data center, we show that our system can be inexpensively deployed in an online setting, delivering low latency when serving users at scale.

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="1512.02595")

Code

Syntology Ran 2 of 39 code samples harvested from 10 repositories linked to this paper; 37 have no recorded run. Of those that ran: 1 ran · violated contract; 1 ran · our draft was wrong.

By repository: community (archive-listed): 39 samples from 10 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.

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

PaddlePaddle/PaddleSpeech officialpaddleApache-2.0 report
2023-MindSpore-1/ms-code-57 mentioned on GitHubmindspore report
2023-MindSpore-1/ms-code-63 mentioned on GitHubmindspore report
DeepMark/deepmark mentioned on GitHubtorchApache-2.0 report
GavinGuan95/Punctuator.Pytorch mentioned on GitHubpytorch report
MangoMoe/VerbalVim mentioned on GitHubpytorch report
PaddlePaddle/models mentioned on GitHubpaddleApache-2.0 report
SeanNaren/deepspeech.pytorch mentioned on GitHubpytorchMIT report
SeanNaren/deepspeech.torch mentioned on GitHubtorchMIT report
TensorSpeech/TensorFlowASR mentioned on GitHubtfApache-2.0 report
UnofficialJuliaMirror/DeepMark-deepmark mentioned on GitHubtorchApache-2.0 report
UnofficialJuliaMirrorSnapshots/DeepMark-deepmark mentioned on GitHubtorchApache-2.0 report
baidu-research/warp-ctc mentioned on GitHubtorchApache-2.0 report
cosmoquester/speech-recognition mentioned on GitHubtfMIT report
fd873630/deep_speech_2_korean mentioned on GitHubpytorchMIT report
freshtan/deepspeech2 mentioned on GitHubmindsporeApache-2.0 report
hkakitani/deepspeech.pytorch mentioned on GitHubpytorchMIT report
mangelroman/audio2score mentioned on GitHubpytorchMIT report
msalhab96/SpeeQ mentioned on GitHubpytorch report
myrtleSoftware/deepspeech mentioned on GitHubpytorchNOASSERTION report
raraz15/DeepTurkish mentioned on GitHubpytorchMIT report
robmsmt/KerasDeepSpeech mentioned on GitHubtfAGPL-3.0 report
sooftware/End-to-End-Speech-Recognition-Models mentioned on GitHubpytorchApache-2.0 report
sooftware/OpenSpeech mentioned on GitHubpytorchMIT report
switiz/deepspeech2.pytorch mentioned on GitHubpytorchApache-2.0 report
tensorflow/models mentioned on GitHubtf report
gitlab.com/sburud/master mentioned on GitHubpytorch report
wangys102/MindSpore mindsporeApache-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

39 samples harvested; 2 ran; 0 honoured the contract we drafted; 37 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.

1ran · violated contract
1ran · our draft was wrong
37unverified

Licence: 0 of the 39 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 10 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.

filter_short_utterances GavinGuan95/Punctuator.Pytorch/data/ted.py community (archive-listed) ran · violated contract MIT (permissive) · 9e067daa4caf2798 · report
get_utterances_from_stm GavinGuan95/Punctuator.Pytorch/data/ted.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 82d083fca2fe6ff4 · report
analyze_words raraz15/DeepTurkish/utilities/text_format.py community (archive-listed) unverified MIT (permissive) · 10e460eab230057c · report
audiofile_to_input_vector bjtommychen/Keras_DeepSpeech2_SpeechRecognition/util/audio.py community (archive-listed) unverified MIT (permissive) · 988e98b189398316 · report
bracket_filter fd873630/deep_speech_2_korean/pre_processing_for_AI_hub_data.py community (archive-listed) unverified MIT (permissive) · 8c10b7268fa78056 · report
calculate_cer mangelroman/audio2score/utils.py community (archive-listed) unverified MIT (permissive) · 32429a57833e5b21 · report
calculate_wer mangelroman/audio2score/utils.py community (archive-listed) unverified MIT (permissive) · 3a1a5ae272c7dfdb · report
cer switiz/deepspeech2.pytorch/utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 5eeba5cf42081f13 · report
check_envirionment switiz/deepspeech2.pytorch/utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 5cb79c0f243a5040 · report
check_loss SeanNaren/deepspeech.pytorch/deepspeech_pytorch/utils.py community (archive-listed) unverified MIT (permissive) · ca014346b646227a · report
clean_text raraz15/DeepTurkish/utilities/text_format.py community (archive-listed) unverified MIT (permissive) · 1486f8a5a335a589 · report
create_train_val_partition raraz15/DeepTurkish/model/data_loader.py community (archive-listed) unverified MIT (permissive) · 0a6358ab2d2cde39 · report
ctcBeamSearch raraz15/DeepTurkish/decoders/BeamSearch.py community (archive-listed) unverified MIT (permissive) · f764213dd1f6bff8 · report
ctc_lambda_func bjtommychen/Keras_DeepSpeech2_SpeechRecognition/Keras_DS2_SpeechRecognition.py community (archive-listed) unverified MIT (permissive) · b671f5730f946b3d · report
emptyCache raraz15/DeepTurkish/decoders/Loss.py community (archive-listed) unverified MIT (permissive) · 97dd44510f634ab4 · report
extendByBlanks raraz15/DeepTurkish/decoders/Common.py community (archive-listed) unverified MIT (permissive) · d5863fd24a880e87 · report
filenum_padding fd873630/deep_speech_2_korean/make_korea_json_label.py community (archive-listed) unverified MIT (permissive) · feb2e0517d7590e7 · report
filter_sentence_list raraz15/DeepTurkish/utilities/text_format.py community (archive-listed) unverified MIT (permissive) · 36b307919659b64f · report
get_lr freshtan/deepspeech2/src/lr_generator.py community (archive-listed) unverified Apache-2.0 (permissive) · 2adf2c08a0fbed04 · report
get_path fd873630/deep_speech_2_korean/pre_processing_for_AI_hub_data.py community (archive-listed) unverified MIT (permissive) · 67b46d997a6d2d02 · report
is_valid_poly wangys102/MindSpore/east/detect.py community (archive-listed) unverified Apache-2.0 (permissive) · 22b8e8ecc724c6f4 · report
load_dataset switiz/deepspeech2.pytorch/utils.py community (archive-listed) unverified Apache-2.0 (permissive) · 8210be28ca690ac0 · report
loadfromecsv bjtommychen/Keras_DeepSpeech2_SpeechRecognition/Keras_DS2_SpeechRecognition.py community (archive-listed) unverified MIT (permissive) · c9ba0819dfef9bee · report
make_loaders raraz15/DeepTurkish/model/data_loader.py community (archive-listed) unverified MIT (permissive) · 612369e6364fb914 · report
multiply_probs raraz15/DeepTurkish/decoders/BeamSearch.py community (archive-listed) unverified MIT (permissive) · 6d9a921d68e2df89 · report
pad_collate_fn raraz15/DeepTurkish/model/data_loader.py community (archive-listed) unverified MIT (permissive) · e32b0e8512dc35f0 · report
pad_list mangelroman/audio2score/utils.py community (archive-listed) unverified MIT (permissive) · ce10a394eb6dd5f4 · report
parseIntList mangelroman/audio2score/prepare.py community (archive-listed) unverified MIT (permissive) · 7ddd1a77d6823719 · report
parseList mangelroman/audio2score/prepare.py community (archive-listed) unverified MIT (permissive) · da8541fdf204f0e1 · report
read_txt raraz15/DeepTurkish/decoders/LanguageModel.py community (archive-listed) unverified MIT (permissive) · be421821ea5166fc · report
recLabelingProb raraz15/DeepTurkish/decoders/Loss.py community (archive-listed) unverified MIT (permissive) · 7859a5f3ad787675 · report
resize_img wangys102/MindSpore/east/detect.py community (archive-listed) unverified Apache-2.0 (permissive) · 27298601d9541728 · report
run_evaluation hkakitani/deepspeech.pytorch/deepspeech_pytorch/testing.py community (archive-listed) unverified MIT (permissive) · d954560f1b1b7360 · report
sparse_tensor_value_to_texts bjtommychen/Keras_DeepSpeech2_SpeechRecognition/util/text.py community (archive-listed) unverified MIT (permissive) · 7970c223f17766e6 · report
sparse_tuple_from bjtommychen/Keras_DeepSpeech2_SpeechRecognition/util/text.py community (archive-listed) unverified MIT (permissive) · 796998505f97f73c · report
sum_probs raraz15/DeepTurkish/decoders/BeamSearch.py community (archive-listed) unverified MIT (permissive) · 5e0550777bebd5b9 · report
text_to_char_array bjtommychen/Keras_DeepSpeech2_SpeechRecognition/util/text.py community (archive-listed) unverified MIT (permissive) · 752dbb28ff7a8b3a · report
to_np hkakitani/deepspeech.pytorch/deepspeech_pytorch/logger.py community (archive-listed) unverified MIT (permissive) · 2e70840fa1254dee · report
wordToLabelSeq raraz15/DeepTurkish/decoders/Common.py community (archive-listed) unverified MIT (permissive) · 9a614a18231dc307 · report

Tasks

Accented Speech RecognitionNoisy Speech RecognitionSpeech Recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Accented Speech Recognition VoxForge American-Canadian Deep Speech 2 Percentage error 7.55 #1 of 2 Archive leaderboard report
Accented Speech Recognition VoxForge Commonwealth Deep Speech 2 Percentage error 13.56 #1 of 2 Archive leaderboard report
Accented Speech Recognition VoxForge European Deep Speech 2 Percentage error 17.55 #1 of 2 Archive leaderboard report
Accented Speech Recognition VoxForge Indian Deep Speech 2 Percentage error 22.44 #1 of 2 Archive leaderboard report
Noisy Speech Recognition CHiME clean Deep Speech 2 Percentage error 3.34 #1 of 2 Archive leaderboard report
Noisy Speech Recognition CHiME real Deep Speech 2 Percentage error 21.79 #4 of 5 Archive leaderboard report
Speech Recognition LibriSpeech test-clean Deep Speech 2 Word Error Rate (WER) 5.33 #57 of 64 Archive leaderboard report
Speech Recognition LibriSpeech test-other Deep Speech 2 Word Error Rate (WER) 13.25 #50 of 53 Archive leaderboard report
Speech Recognition WSJ eval92 Deep Speech 2 Word Error Rate (WER) 3.60 #14 of 17 Archive leaderboard report
Speech Recognition WSJ eval93 Deep Speech 2 Word Error Rate (WER) 4.98 #1 of 3 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.

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