Papers › Robust Speech Recognition via Large-Scale Weak Supervision

Robust Speech Recognition via Large-Scale Weak Supervision

6 Dec 2022Preprint 2022 9arXiv:2212.04356archive 2025-07-28

Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever

We study the capabilities of speech processing systems trained simply to predict large amounts of transcripts of audio on the internet. When scaled to 680,000 hours of multilingual and multitask supervision, the resulting models generalize well to standard benchmarks and are often competitive with prior fully supervised results but in a zero-shot transfer setting without the need for any fine-tuning. When compared to humans, the models approach their accuracy and robustness. We are releasing models and inference code to serve as a foundation for further work on robust speech processing.

PaperPDFConference PDFCodeCode Syntology ran

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):

Code

No sample from the archive's official repository (openai/whisper) is linked to this paper; the 59 samples below come from 10 community repositories. Syntology harvested code from openai/whisper for another paper.

openai/whisper: Syntology's graph links these samples to OmniTrace: A Unified Framework for Generation-Time Attribution in Omni-Modal LLMs (arXiv:2604.13073; that paper's own run record: 18 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 5 honoured, 3 violated, 0 with no contract checked; 10 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method.

Syntology 59 code samples harvested from 10 repositories linked to this paper: 49 ran (of which 0 constructed an object rather than computing a result; 36 with no instrument failure: 4 honoured, 1 violated, 31 with no contract checked; 13 where Syntology's instrument failed) · 10 unverified. Where Syntology's instrument failed: 11 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 59 samples from 10 repositories, 49 ran (36 with no instrument failure; 13 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 5 of its repositories that Syntology harvested for other papers:

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. 1 of the 49 samples that ran has no recorded date.

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

openai/whisper officialmentioned in papermentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to OmniTrace: A Unified Framework for Generation-Time Attribution in Omni-Modal LLMs (arXiv:2604.13073; that paper's own run record: 18 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 5 honoured, 3 violated, 0 with no contract checked; 10 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
audioshake/alt-eval mentioned on GitHubMITSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
briansidp/whisperbiasing mentioned on GitHubpytorchMITSyntology: 13 samples, 10 ran (3 instrument-failed) · 3 unverified report
collabora/whisperlive mentioned on GitHubpytorchMITSyntology: 2 samples, 1 ran (0 instrument-failed) · 1 unverified report
collabora/whisperspeech mentioned on GitHubpytorchMITSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
k2-fsa/icefall mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Unifying Streaming and Non-streaming Zipformer-based ASR (arXiv:2506.14434; that paper's own run record: 11 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 8 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
kadirnar/whisper-plus mentioned on GitHubpytorchApache-2.0Syntology: 7 samples, 4 ran (0 instrument-failed) · 3 unverified report
open-creator/icefall mentioned on GitHubpytorchApache-2.0Syntology: 11 samples, 9 ran (6 instrument-failed) · 2 unverified report
robflynnyh/long-context-asr mentioned on GitHubpytorchApache-2.0Syntology: 10 samples, 10 ran (0 instrument-failed) · 0 unverified report
sanchit-gandhi/whisper-jax mentioned on GitHubjaxApache-2.0Syntology: 4 samples, 4 ran (1 instrument-failed) · 0 unverified report
whisperspeech/whisperspeech mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
ggerganov/whisper.cpp MITSyntology: 7 samples, 6 ran (1 instrument-failed) · 1 unverified report
huggingface/transformers pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Attention Is All You Need (arXiv:1706.03762; that paper's own run record: 610 ran (of which 293 constructed an object rather than computing a result; 529 with no instrument failure: 45 honoured, 15 violated, 469 with no contract checked; 81 where Syntology's instrument failed) · 336 unverified), HuggingFace's Transformers: State-of-the-art Natural Language Processing (arXiv:1910.03771; that paper's own run record: 3 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; 3 where Syntology's instrument failed) · 0 unverified), DeBERTa: Decoding-enhanced BERT with Disentangled Attention (arXiv:2006.03654; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 9 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
m-bain/whisperx pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to WhisperX: Time-Accurate Speech Transcription of Long-Form Audio (arXiv:2303.00747; that paper's own run record: 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); nothing checks that they implement this paper's method. report
pwc-1/Paper-9 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RandAugment: Practical automated data augmentation with a reduced search space (arXiv:1909.13719; that paper's own run record: 58 ran (of which 1 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 51 where Syntology's instrument failed) · 7 unverified); nothing checks that they implement this paper's method. 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.

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

59 samples harvested: 49 ran (of which 0 constructed an object rather than computing a result; 36 with no instrument failure: 4 honoured, 1 violated, 31 with no contract checked; 13 where Syntology's instrument failed) · 10 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

4ran · honoured contract
1ran · violated contract
11ran · our draft was wrong
2ran; the contract check could not run
31ran
10unverified

Licence: 21 of the 59 samples are pointer only, meaning Syntology does not serve that copy's text. For 3 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. 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. 1 sample whose link to this paper records no repository is listed under one of this paper's own repositories, because identical code was harvested from it for another 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.

audioshake/alt-eval community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
normalize_lyrics src/alt_eval/normalization.py ran fingerprinted MIT (permissive) · c4ee356de1bd8c77 · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-25
to_rich_tokens src/alt_eval/tokenizer.py ran MIT (permissive) · 2ef9c9c7ee284aaf · report fixture: signature tokens:list/1/str · returned list · level recorded 2026-09-25
tokens_as_words src/alt_eval/tokenizer.py ran MIT (permissive) · 3d1de040cc006f6a · report fixture: signature tokens:list/1/Token · returned list · level recorded 2026-09-25
briansidp/whisperbiasing community (archive-listed) 13 samples, 10 ran (3 instrument-failed) · 3 unverified
check_in_utt dataloader.py ran MIT recorded; this copy not marked cleared · pointer only · fa47326ca5e0886b · report fixture: signature tok_word:list/1/int, target:list/1/int · returned bool · level recorded 2026-09-25
collate_wrapper in 5 papers dataloader.py ran MIT recorded; this copy not marked cleared · pointer only · c4c3b54ad53b1fce · report fixture: signature batch:list/1/tuple · returned tuple · level recorded 2026-09-25
exact_div in 9 papers whisper/utils.py ran · honoured contract fingerprinted MIT recorded; this copy not marked cleared · pointer only · 6e0808fc4828513a · report fixture: signature x:int/0/int, y:int/0/int · returned int · level recorded 2026-09-22
make_lexical_tree dataloader.py ran MIT recorded; this copy not marked cleared · pointer only · c45f3b632ec36534 · report fixture: signature word_dict:dict/0/?, subword_dict:dict/0/?, word_unk:int/0/int · returned list · level recorded 2026-09-25
optional_int in 5 papers whisper/utils.py ran · honoured contract MIT recorded; this copy not marked cleared · pointer only · e56996b20b9c19cb · report fixture: signature string:str/0/str · returned int · level recorded 2026-09-25
pad_or_trim in 12 papers whisper/audio.py ran · our draft was wrong MIT recorded; this copy not marked cleared · pointer only · 9f13de1c23e4003d · report fixture: signature array:Tensor/2/float, length:int/0/int, axis:int/0/int · returned Tensor · level recorded 2026-09-25
remove_symbols whisper/normalizers/basic.py ran · our draft was wrong fingerprinted MIT recorded; this copy not marked cleared · pointer only · 2206d84b6900ab3e · report fixture: signature s:str/0/str · returned str · level recorded 2026-09-25
remove_symbols_and_diacritics whisper/normalizers/basic.py ran · our draft was wrong fingerprinted MIT recorded; this copy not marked cleared · pointer only · cef27f679a755971 · report fixture: signature s:str/0/str, keep:str/0/str · returned str · level recorded 2026-09-25
sinusoids in 13 papers whisper/model.py ran · honoured contract fingerprinted MIT recorded; this copy not marked cleared · pointer only · e529c9641c178fe0 · report fixture: signature length:int/0/int, channels:int/0/int, max_timescale:int/0/int · returned Tensor · level recorded 2026-09-22
str2bool in 533 papers whisper/utils.py ran · violated contract MIT recorded; this copy not marked cleared · pointer only · 176b6ba6c2c2940c · report fixture: signature string:str/0/str · returned bool · level recorded 2026-09-25
build_tokenizer in 19 papers whisper/tokenizer.py unverified MIT recorded; this copy not marked cleared · pointer only · acb6a55d43c7f465 · report
mel_filters in 9 papers whisper/audio.py unverified MIT recorded; this copy not marked cleared · pointer only · edf924293b4b2375 · report
transcribe in 8 papers whisper/transcribe.py unverified MIT recorded; this copy not marked cleared · pointer only · 738c6e74fd16381d · report
collabora/whisperlive community (archive-listed) 2 samples, 1 ran (0 instrument-failed) · 1 unverified
format_time in 140 papers whisper_live/utils.py ran fingerprinted MIT (permissive) · 9649e15864c56627 · report fixture: signature s:float/0/float · returned str · level recorded 2026-09-25
load_audio in 10 papers whisper_live/diarization.py unverified MIT (permissive) · 34dd5baae0e5830e · report
collabora/whisperspeech community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
multinomial_sample_one_no_sync in 10 papers whisperspeech/inference.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · our draft was wrong fingerprinted licence of this copy not recorded · 0b5f90e8856ad7af · report fixture: signature probs_sort:Tensor/2/float · returned Tensor · level recorded, date not recorded
kadirnar/whisper-plus community (archive-listed) 7 samples, 4 ran (0 instrument-failed) · 3 unverified
compression_ratio whisperplus/pipelines/lightning_whisper_mlx/decoding.py ran fingerprinted Apache-2.0 (permissive) · 5632d6609129dfa3 · report fixture: signature text:str/0/str · returned float · level recorded 2026-09-25
dtw_cpu whisperplus/pipelines/lightning_whisper_mlx/timing.py ran · honoured contract Apache-2.0 (permissive) here; the stored copy is pointer only · 76330ed965e60af5 · report fixture: returned ndarray · level recorded 2026-09-25
median_filter in 7 papers whisperplus/pipelines/lightning_whisper_mlx/timing.py ran Apache-2.0 (permissive) · 6f669ba80cbbf823 · report fixture: signature x:ndarray/3/float, filter_width:int/0/int · returned ndarray · level recorded 2026-09-25
sinusoids in 13 papers whisperplus/pipelines/lightning_whisper_mlx/torch_whisper.py ran fingerprinted Apache-2.0 (permissive) · 407b4297398ff29d · report fixture: signature length:int/0/int, channels:int/0/int, max_timescale:int/0/int · returned Tensor · level recorded 2026-09-25
backtrace whisperplus/pipelines/lightning_whisper_mlx/timing.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 41c654294ba9ec56 · report
get_encoding whisperplus/pipelines/lightning_whisper_mlx/tokenizer.py unverified Apache-2.0 (permissive) · b3916c1b2edcf996 · report
get_tokenizer in 156 papers whisperplus/pipelines/lightning_whisper_mlx/tokenizer.py unverified Apache-2.0 (permissive) · f2f52f6a90834243 · report
open-creator/icefall community (archive-listed) 11 samples, 9 ran (6 instrument-failed) · 2 unverified
byte_decode icefall/byte_utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 59d410aa19bdef96 · report fixture: signature x:str/0/str · returned str · level recorded 2026-09-25
byte_encode icefall/byte_utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 004eb0067bd43754 · report fixture: signature x:str/0/str · returned str · level recorded 2026-09-25
convert_alignments_to_tensor icefall/ali.py ran · our draft was wrong Apache-2.0 (permissive) · 8a64b592c82c0d10 · report fixture: signature alignments:dict/0/?, device:device/0/? · returned dict · level recorded 2026-09-25
find_checkpoints icefall/checkpoint.py ran · our draft was wrong Apache-2.0 (permissive) · 87c9b24deaf27799 · report fixture: signature out_dir:PosixPath/0/?, iteration:int/0/int · returned list · level recorded 2026-09-26
get_class_name in 11 papers icefall/diagnostics.py ran · our draft was wrong Apache-2.0 (permissive) · b865558252e1566a · report fixture: signature module:callable/0/? · returned str · level recorded 2026-09-26
get_tensor_stats icefall/diagnostics.py ran Apache-2.0 (permissive) · d1162925e2d53784 · report fixture: signature x:Tensor/4/float, dim:int/0/int, stats_type:str/0/str · returned tuple · level recorded 2026-09-25
load_alignments icefall/ali.py ran Apache-2.0 (permissive) · a972a5df2771f7a5 · report fixture: signature filename:str/0/str · returned tuple · level recorded 2026-09-25
lookup_alignments icefall/ali.py ran Apache-2.0 (permissive) · a1f5678203ff47e8 · report fixture: signature cut_ids:list/1/str, alignments:dict/0/?, num_classes:int/0/int, log_score:float/0/float · returned Tensor · level recorded 2026-09-25
smart_byte_decode icefall/byte_utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 226350299950338b · report fixture: signature x:str/0/str · returned str · level recorded 2026-09-25
attach_diagnostics icefall/diagnostics.py unverified Apache-2.0 (permissive) · 5969d689c8a84ef2 · report
average_checkpoints in 21 papers icefall/checkpoint.py unverified Apache-2.0 (permissive) · 69124b03f4b749bd · report
robflynnyh/long-context-asr community (archive-listed) 10 samples, 10 ran (0 instrument-failed) · 0 unverified
LinearActivation in 16 papers lcasr/components/long_conv.py ran Apache-2.0 (permissive) · 6ecbcf4f3d00f7ed · report fixture: signature d_input:int/0/int, d_output:int/0/int, bias:bool/0/bool, zero_bias_init:bool/0/bool, transposed:bool/0/bool, initializer:str/0/str, activation:str/0/str, activate:bool/0/bool, weight_norm:bool/0/bool · returned Sequential · level recorded 2026-09-25
blank_p exp/train_utterance_meta_loss_pred.py ran Apache-2.0 (permissive) · 3787ac531561b6c6 · report fixture: signature logits:Tensor/3/float, tokenizer:MockTokenizer/0/? · returned float · level recorded 2026-09-25
blank_p exp/train_utterance_meta_loss_pred_ce.py ran Apache-2.0 (permissive) · 9f16788e5da13dfa · report fixture: signature logits:Tensor/3/float, tokenizer:MinimalTokenizer/0/? · returned float · level recorded 2026-09-25
check_if_already_evaluated eval/eval_manager.py ran Apache-2.0 (permissive) · 9140147b0fd2c440 · report fixture: signature model_save_path:str/0/str, cur_df:DataFrame/0/?, dataset:str/0/str, split:str/0/str, args:Args/0/? · returned bool · level recorded 2026-09-25
get_args in 84 papers eval/eval_manager.py ran Apache-2.0 (permissive) · 63302892207b7f5b · report fixture: signature config:dict/0/?, split:str/0/str, model:callable/0/?, dataset_config:dict/0/? · returned ArgsClass · level recorded 2026-09-25
get_data_to_save eval/eval_manager.py ran Apache-2.0 (permissive) · ae28fa768d63729b · report fixture: signature config:MockConfig/0/?, wers:list/1/dict, split:str/0/str, dataset:str/0/str, model:callable/0/? · returned list · level recorded 2026-09-25
get_dtype in 14 papers exp/train_utterance_meta_loss_pred.py ran Apache-2.0 (permissive) · ad63708b6da385d5 · report fixture: signature dtype:str/0/str · returned dtype · level recorded 2026-09-25
get_initializer in 32 papers lcasr/components/long_conv.py ran Apache-2.0 (permissive) · ee3c8d2a605ac8bc · report fixture: signature name:str/0/str, activation:str/0/str · returned partial · level recorded 2026-09-25
laplace in 12 papers lcasr/components/long_conv.py ran fingerprinted Apache-2.0 (permissive) · a799a46446fafd99 · report fixture: signature x:Tensor/2/float, mu:float/0/float, sigma:float/0/float · returned Tensor · level recorded 2026-09-25
rgetattr in 22 papers exp/run_launcher.py ran Apache-2.0 (permissive) · bdbf18d0ee9006f7 · report fixture: signature obj:callable/0/?, attr:str/0/str · returned int · level recorded 2026-09-25
sanchit-gandhi/whisper-jax community (archive-listed) 4 samples, 4 ran (1 instrument-failed) · 0 unverified
dot_product_attention in 15 papers whisper_jax/layers.py ran; the contract check could not run (RAISES) Apache-2.0 recorded; this copy not marked cleared · pointer only · 2000f34f95c683d6 · report fixture: signature query:ArrayImpl/4/float, key:ArrayImpl/4/float, value:ArrayImpl/4/float, bias:ArrayImpl/4/float, dropout_rng:ArrayImpl/1/int, dropout_rate:float/0/float, deterministic:bool/0/bool, dtype:callable/0/?, float32_logits:bool/0/bool · returned ArrayImpl · level recorded 2026-09-24
pjit whisper_jax/partitioner.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 87bdf3970749dc11 · report fixture: signature fun:callable/0/?, in_axis_resources:tuple/1/NoneType, out_axis_resources:NoneType/0/?, static_argnums:tuple/1/empty, donate_argnums:tuple/1/empty, backend:NoneType/0/? · returned PjitFunction · level recorded 2026-09-25
pjit_with_cpu_fallback whisper_jax/partitioner.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · d35e5ea9b14ce05e · report fixture: signature fun:callable/0/?, in_axis_resources:tuple/1/NoneType, out_axis_resources:NoneType/0/?, static_argnums:tuple/1/empty, donate_argnums:tuple/1/empty, backend:NoneType/0/? · returned PjitFunction · level recorded 2026-09-25
with_sharding_constraint whisper_jax/partitioner.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · e5a9a9806511efb5 · report fixture: signature x:ArrayImpl/2/float, axis_resources:NoneType/0/? · returned ArrayImpl · level recorded 2026-09-25
whisperspeech/whisperspeech community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
logits_to_probs in 18 papers whisperspeech/inference.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · 8a8edc42e5e25cea · report fixture: signature logits:Tensor/2/float, T:float/0/float, top_k:int/0/int · returned Tensor · level recorded 2026-09-22
ggerganov/whisper.cpp community (archive-listed) 7 samples, 6 ran (1 instrument-failed) · 1 unverified
correct_for_bias_scale_order_inversion models/convert-whisper-to-coreml.py ran MIT (permissive) · b41e987fee8a92ff · report fixture: signature state_dict:dict/0/?, prefix:str/0/str, local_metadata:dict/0/?, strict:bool/0/bool, missing_keys:list/1/empty, unexpected_keys:list/1/empty, error_msgs:list/1/empty · returned dict · level recorded 2026-09-25
get_short_name in 6 papers ggml/src/ggml-cuda/template-instances/generate_cu_files.py ran · our draft was wrong fingerprinted MIT (permissive) here; the stored copy is pointer only · a5b8c22a3ac7b238 · report fixture: signature long_quant_name:str/0/str · returned str · level recorded 2026-09-25
hz_to_mel models/convert-parakeet-to-ggml.py ran fingerprinted MIT (permissive) · 9720e570a35f6d55 · report fixture: signature freq:ndarray/1/float · returned ndarray · level recorded 2026-09-25
load_model_config in 8 papers models/convert-parakeet-to-ggml.py ran MIT (permissive) · ba1be9f251ecd80a · report fixture: signature config_path:str/0/str · returned dict · level recorded 2026-09-25
mel_to_hz models/convert-parakeet-to-ggml.py ran fingerprinted MIT (permissive) · 40e421c01b760f1c · report fixture: signature mel:float/0/float · returned float · level recorded 2026-09-25
rename_keys models/convert-h5-to-coreml.py ran MIT (permissive) · e65744d3337badd5 · report fixture: signature s_dict:dict/0/? · returned dict · level recorded 2026-09-25

Tasks

Robust Speech RecognitionSpeech RecognitionSpeech-to-Speech TranslationZero-Shot Audio Retrievalspeech-recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Recognition Common Voice English Whisper (Large v2) Word Error Rate (WER) 9.4% #2 of 2 Archive leaderboard report
Speech Recognition Common Voice French Whisper (Large v2) Test WER 13.9% #8 of 8 Archive leaderboard report
Speech Recognition Common Voice German Whisper (Large v2) Test WER 6.4% #7 of 14 Archive leaderboard report
Speech Recognition Common Voice Italian Whisper (Large v2) Test WER 7.1% #1 of 2 Archive leaderboard report
Speech Recognition Common Voice Japanese Whisper (Large v2) Test WER 9.1% #1 of 1 Archive leaderboard report
Speech Recognition Common Voice Russian Whisper (Large v2) Test WER 7.1% #1 of 1 Archive leaderboard report
Speech Recognition Common Voice Spanish Whisper (Large v2) Test WER 5.6% #2 of 8 Archive leaderboard report
Speech-to-Speech Translation FLEURS X-eng WhisperV2 ASR-BLEU 23.5 #6 of 7 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