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
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.
In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.
Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
For agents, on Syntology's MCP service (how to connect):
get_harvested_code_for_paper(arxiv_id="1512.02595")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="1512.02595", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="1512.02595")
Code
No sample from the archive's official repository (PaddlePaddle/PaddleSpeech) is linked to this paper; the 39 samples below come from 10 community repositories. Syntology harvested code from PaddlePaddle/PaddleSpeech for another paper.
PaddlePaddle/PaddleSpeech: Syntology's graph links these samples to A³T: Alignment-Aware Acoustic and Text Pretraining for Speech Synthesis and Editing (arXiv:2203.09690; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method.
Syntology 39 code samples harvested from 10 repositories linked to this paper: 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. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.
By repository: community (archive-listed): 39 samples from 10 repositories, 33 ran (32 with no instrument failure; 1 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.
No sample is linked to this paper from 7 of its repositories that Syntology harvested for other papers:
- PaddlePaddle/PaddleSpeech, official repository: harvested for another paper, listed under the repository
- PaddlePaddle/models, community (archive-listed): harvested for another paper, listed under the repository
- TensorSpeech/TensorFlowASR, community (archive-listed): harvested for another paper, listed under the repository
- cosmoquester/speech-recognition, community (archive-listed): harvested for another paper, listed under the repository
- msalhab96/SpeeQ, community (archive-listed): harvested for another paper, listed under the repository
- tensorflow/models, community (archive-listed): harvested for 4 other papers, listed under the repository
- mindspore-ai/models, community (archive-listed): harvested for another paper, listed under the repository
Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.
35 repositories listed; official and paper-mentioned ones first.
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.
A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.
Code Syntology ran Syntology
39 samples harvested: 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. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.
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; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.
Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.
A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.
9e067daa4caf2798 · report
fixture: signature utterance_info:dict/0/?, min_len_sec:float/0/float · returned bool · level recorded 2026-09-22 82d083fca2fe6ff4 · report
fixture: signature stm_file:str/0/str · returned list · level recorded 2026-09-22 ca014346b646227a · report
fixture: signature loss:Tensor/1/float, loss_value:float/0/float · returned tuple · level recorded 2026-09-26 c9ba0819dfef9bee · report
fixture: signature fname:str/0/str · returned list · level recorded 2026-09-26 7970c223f17766e6 · report
fixture: signature value:MockSparseTensor/0/?, alphabet:MockAlphabet/0/? · returned list · level recorded 2026-09-26 752dbb28ff7a8b3a · report
fixture: signature original:str/0/str, alphabet:Alphabet/0/? · returned ndarray · level recorded 2026-09-26 b671f5730f946b3d · report
8c10b7268fa78056 · report
fixture: signature sentence:str/0/str · returned str · level recorded 2026-09-26 feb2e0517d7590e7 · report
fixture: signature filenum:int/0/int · returned str · level recorded 2026-09-26 67b46d997a6d2d02 · report
fixture: signature path:str/0/str, fname:str/0/str, filenum:str/0/str, format:str/0/str · returned str · level recorded 2026-09-26 2adf2c08a0fbed04 · report
fixture: signature lr_init:float/0/float, total_epochs:int/0/int, steps_per_epoch:int/0/int · returned ndarray · level recorded 2026-09-26 d954560f1b1b7360 · report
fixture: signature test_loader:list/1/tuple, device:device/0/?, model:callable/0/?, decoder:DummyDecoder/0/?, target_decoder:DummyTargetDecoder/0/?, save_output:NoneType/0/?, verbose:bool/0/bool, use_half:bool/0/bool · returned tuple · level recorded 2026-09-26 census: 8 implementations from 14 papers, 2 distinct outputs across 2 buckets, one shared input per bucket deepspeech_pytorch/logger.py ran MIT (permissive) ·
2e70840fa1254dee · report
fixture: signature x:Tensor/2/float · returned ndarray · level recorded 2026-09-24 32429a57833e5b21 · report
fixture: signature s1:str/0/str, s2:str/0/str, word_sep:str/0/str · returned tuple · level recorded 2026-09-26 3a1a5ae272c7dfdb · report
fixture: signature s1:str/0/str, s2:str/0/str, word_sep:str/0/str · returned tuple · level recorded 2026-09-26 ce10a394eb6dd5f4 · report
fixture: signature xs:list/1/Tensor, pad_value:float/0/float · returned Tensor · level recorded 2026-09-25 7ddd1a77d6823719 · report
fixture: signature string:str/0/str · returned list · level recorded 2026-09-26 da8541fdf204f0e1 · report
fixture: signature string:str/0/str · returned list · level recorded 2026-09-26 10e460eab230057c · report
fixture: signature sentences:list/1/str · returned Counter · level recorded 2026-09-26 1486f8a5a335a589 · report
fixture: signature text:str/0/str · returned str · level recorded 2026-09-26 0a6358ab2d2cde39 · report
fixture: signature df:DataFrame/0/?, split_ratio:float/0/float, N_batch:int/0/int · returned tuple · level recorded 2026-09-26 97dd44510f634ab4 · report
fixture: signature maxT:int/0/int, labelingWithBlanks:list/1/int · returned list · level recorded 2026-09-26 d5863fd24a880e87 · report
fixture: signature seq:list/1/int, b:int/0/int · returned list · level recorded 2026-09-26 36b307919659b64f · report
fixture: signature sentence_list:list/1/str, clean_sentences:list/1/empty, bound:int/0/int · returned list · level recorded 2026-09-26 6d9a921d68e2df89 · report
fixture: signature ln_pa:Tensor/2/float, ln_pb:Tensor/2/float · returned Tensor · level recorded 2026-09-26 e32b0e8512dc35f0 · report
fixture: signature data:list/1/tuple · returned tuple · level recorded 2026-09-26 be421821ea5166fc · report
fixture: signature file_path:str/0/str, encoding:str/0/str · returned list · level recorded 2026-09-26 7859a5f3ad787675 · report
fixture: signature t:int/0/int, s:int/0/int, mat:Tensor/2/float, labelingWithBlanks:list/1/int, blank:int/0/int, cache:list/1/list · returned Tensor · level recorded 2026-09-26 5e0550777bebd5b9 · report
fixture: signature ln_pa:float/0/float, ln_pb:float/0/float · returned float · level recorded 2026-09-26 9a614a18231dc307 · report
fixture: signature w:str/0/str, classes:list/1/str · returned list · level recorded 2026-09-26 5cb79c0f243a5040 · report
fixture: signature use_cuda:bool/0/bool · returned device · level recorded 2026-09-26 8210be28ca690ac0 · report
fixture: signature transcripts_path:str/0/str · returned tuple · level recorded 2026-09-26 22b8e8ecc724c6f4 · report
fixture: signature res:Tensor/2/float, score_shape:tuple/1/int, scale:int/0/int · returned bool · level recorded 2026-09-26 27298601d9541728 · report
fixture: signature img:Image/0/? · returned tuple · level recorded 2026-09-26 Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| 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