Papers › Squeezeformer: An Efficient Transformer for Automatic Speech Recognition

Squeezeformer: An Efficient Transformer for Automatic Speech Recognition

2 Jun 2022arXiv:2206.00888archive 2025-07-28

Sehoon Kim, Amir Gholami, Albert Shaw, Nicholas Lee, Karttikeya Mangalam, Jitendra Malik, Michael W. Mahoney, Kurt Keutzer

The recently proposed Conformer model has become the de facto backbone model for various downstream speech tasks based on its hybrid attention-convolution architecture that captures both local and global features. However, through a series of systematic studies, we find that the Conformer architecture's design choices are not optimal. After re-examining the design choices for both the macro and micro-architecture of Conformer, we propose Squeezeformer which consistently outperforms the state-of-the-art ASR models under the same training schemes. In particular, for the macro-architecture, Squeezeformer incorporates (i) the Temporal U-Net structure which reduces the cost of the multi-head attention modules on long sequences, and (ii) a simpler block structure of multi-head attention or convolution modules followed up by feed-forward module instead of the Macaron structure proposed in Conformer. Furthermore, for the micro-architecture, Squeezeformer (i) simplifies the activations in the convolutional block, (ii) removes redundant Layer Normalization operations, and (iii) incorporates an efficient depthwise down-sampling layer to efficiently sub-sample the input signal. Squeezeformer achieves state-of-the-art results of 7.5%, 6.5%, and 6.0% word-error-rate (WER) on LibriSpeech test-other without external language models, which are 3.1%, 1.4%, and 0.6% better than Conformer-CTC with the same number of FLOPs. Our code is open-sourced and available online.

PaperPDFCodeCode Syntology ranMethod as code

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

Syntology 49 code samples harvested from 3 repositories linked to this paper: 31 ran (of which 12 constructed an object rather than computing a result; 28 with no instrument failure: 3 honoured, 0 violated, 25 with no contract checked; 3 where Syntology's instrument failed) · 18 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 15 samples from 1 repository, 5 ran (4 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 34 samples from 2 repositories, 26 ran (24 with no instrument failure; 2 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 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

kssteven418/squeezeformer officialmentioned in papermentioned on GitHubtfSyntology: 15 samples, 5 ran (1 instrument-failed) · 10 unverified report
msalhab96/SpeeQ mentioned on GitHubpytorchSyntology: 16 samples, 12 ran (1 instrument-failed) · 4 unverified report
upskyy/Squeezeformer mentioned on GitHubpytorchSyntology: 18 samples, 14 ran (1 instrument-failed) · 4 unverified report
NVIDIA/NeMo pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Cross-Modal Robustness Transfer (CMRT): Training Robust Speech Translation Models... (arXiv:2602.11933; that paper's own run record: 5 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; 5 where Syntology's instrument failed) · 1 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

49 samples harvested: 31 ran (of which 12 constructed an object rather than computing a result; 28 with no instrument failure: 3 honoured, 0 violated, 25 with no contract checked; 3 where Syntology's instrument failed) · 18 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.

3ran · honoured contract
3ran · our draft was wrong
25ran
18unverified

Licence: 0 of the 49 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 3 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”. For 8 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

kssteven418/squeezeformer official repository 15 samples, 5 ran (1 instrument-failed) · 10 unverified
FFModule src/models/conformer_encoder.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 29208c0089f89df9 · report fixture: signature inputs:EagerTensor/3/float, training:bool/0/bool · returned EagerTensor · level recorded 2026-09-22
GLU in 12 papers src/models/conformer_encoder.py ran fingerprinted Apache-2.0 (permissive) · 35285e7d41e8ab62 · report fixture: signature inputs:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-24
IdentityLayer in 7 papers src/models/conformer_encoder.py ran fingerprinted Apache-2.0 (permissive) · 4d0d22b07f8f313e · report fixture: signature inputs:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-24
PositionalEncoding in 126 papers
census: 13 implementations from 9 papers, 9 distinct outputs across 5 buckets, one shared input per bucket
src/models/conformer_encoder.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 8008c17ae543b9f8 · report fixture: signature inputs:EagerTensor/3/float · returned EagerTensor · level recorded 2026-09-22
shape_list in 9 papers src/models/conformer_encoder.py ran · our draft was wrong Apache-2.0 (permissive) · f1f0e7aba532e2f6 · report fixture: signature x:EagerTensor/2/float, out_type:DType/0/? · returned list · level recorded 2026-09-22
ConformerBlock src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · c6798a57b40a97c6 · report
ConformerEncoder src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · f8af5a4e29b98e92 · report
Conv2dSubsampling src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · 0c18f86c9e0954b7 · report
ConvFFModule src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · 664e0546c35af991 · report
ConvModule in 8 papers src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · c6e0c0f0739513ec · report
MHSAFFModule src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · 3065ed179d177d1a · report
MHSAModule src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · 556aca63592449fd · report
RelPositionMultiHeadAttention src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · 9350e6b3773fb284 · report
TimeReductionLayer src/models/conformer_encoder.py unverified Apache-2.0 (permissive) · 480648d1fbf231db · report
msalhab96/SpeeQ community (archive-listed) 16 samples, 12 ran (1 instrument-failed) · 4 unverified
AddAndNorm speeq/models/encoders.py ran · metamorphic tier: invariant MIT (permissive) · 17b0a7a2343ae12d · report fixture: signature x:Tensor/3/float, sub_x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
ConformerConvModule speeq/models/encoders.py ran · metamorphic tier: invariant MIT (permissive) · ac9fc9672ce1f61d · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
ConformerFeedForward speeq/models/encoders.py ran · metamorphic tier: invariant MIT (permissive) · e8be6862875569ea · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
Conv1DLayers speeq/models/encoders.py ran MIT (permissive) · 4a641fcb9f0a3313 · report fixture: signature x:Tensor/3/float, data_len:Tensor/1/int · returned tuple · level recorded 2026-09-24
MultiHeadAtt speeq/models/encoders.py ran · metamorphic tier: invariant MIT (permissive) · e3dbd3a46a9bf6f7 · report fixture: signature key:Tensor/3/float, query:Tensor/3/float, value:Tensor/3/float, key_mask:NoneType/0/?, query_mask:NoneType/0/? · returned Tensor · level recorded 2026-09-22
Scaling1d speeq/models/encoders.py ran · metamorphic tier: invariant MIT (permissive) · 8c469852dc208010 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
SqueezeformerConvModule speeq/models/encoders.py ran · metamorphic tier: invariant MIT (permissive) · 4dea36bac3f898c4 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
SqueezeformerFeedForward speeq/models/encoders.py ran · metamorphic tier: invariant MIT (permissive) · e2853e88fa2a4612 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
add_pos_enc speeq/models/encoders.py ran · honoured contract fingerprinted MIT (permissive) · cfc657760fd4bb85 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
calc_data_len speeq/models/encoders.py ran · honoured contract MIT (permissive) · 860c95bcb91d9a54 · report fixture: signature result_len:int/0/int, pad_len:int/0/int, data_len:int/0/int, kernel_size:int/0/int, stride:int/0/int · returned int · level recorded 2026-09-24
get_mask_from_lens speeq/models/encoders.py ran · our draft was wrong fingerprinted MIT (permissive) · c9d2ed75de0f8e2e · report fixture: signature lengths:Tensor/1/int, max_len:int/0/int · returned Tensor · level recorded 2026-09-22
get_positional_encoding in 10 papers speeq/models/encoders.py ran · honoured contract fingerprinted MIT (permissive) · 686e499ae419b99c · report fixture: signature max_length:int/0/int, d_model:int/0/int · returned Tensor · level recorded 2026-09-22
ConformerPreNet speeq/models/encoders.py unverified MIT (permissive) · 77f464c34cf1fc3a · report
SqueezeformerBlock speeq/models/encoders.py unverified MIT (permissive) · 136716ec91fc7e49 · report
SqueezeformerEncoder speeq/models/encoders.py unverified MIT (permissive) · 74f45b8eb78d354f · report
SqueezeformerRelativeMHSA speeq/models/encoders.py unverified MIT (permissive) · 23a9c915a6989d9a · report
upskyy/Squeezeformer community (archive-listed) 18 samples, 14 ran (1 instrument-failed) · 4 unverified
ConvModule in 8 papers squeezeformer/model.py ran Apache-2.0 (permissive) · a6e14d81239e64ad · report fixture: returned ConvModule · level recorded 2026-09-22
DepthwiseConv1d squeezeformer/model.py ran fingerprinted Apache-2.0 (permissive) · f8183fb916aefe76 · report fixture: returned DepthwiseConv1d · level recorded 2026-09-22
DepthwiseConv2d squeezeformer/model.py ran fingerprinted Apache-2.0 (permissive) · 86a8c140aa415e79 · report fixture: returned DepthwiseConv2d · level recorded 2026-09-22
DepthwiseConv2dSubsampling squeezeformer/model.py ran Apache-2.0 (permissive) · 98a41dcbe0265925 · report fixture: signature inputs:Tensor/3/float, input_lengths:Tensor/1/int · returned tuple · level recorded 2026-09-24
FeedForwardModule squeezeformer/model.py ran fingerprinted Apache-2.0 (permissive) · 425b3f6eba4ae90e · report fixture: returned FeedForwardModule · level recorded 2026-09-22
GLU in 12 papers squeezeformer/model.py ran Apache-2.0 (permissive) · 46a7b3e64f094c31 · report fixture: returned GLU · level recorded 2026-09-22
PointwiseConv1d squeezeformer/model.py ran fingerprinted Apache-2.0 (permissive) · 8acc35cd7fb4aa6f · report fixture: returned PointwiseConv1d · level recorded 2026-09-22
RelPositionalEncoding squeezeformer/model.py ran fingerprinted Apache-2.0 (permissive) · f528d6540416bf8b · report fixture: returned RelPositionalEncoding · level recorded 2026-09-22
RelativeMultiHeadAttention squeezeformer/model.py ran Apache-2.0 (permissive) · 0d0315fcda610792 · report fixture: returned RelativeMultiHeadAttention · level recorded 2026-09-22
ResidualConnectionModule squeezeformer/model.py ran fingerprinted Apache-2.0 (permissive) · aabca50483e3eae8 · report fixture: returned ResidualConnectionModule · level recorded 2026-09-22
Swish in 30 papers
census: 22 implementations from 19 papers, 2 distinct outputs on one shared input
squeezeformer/model.py ran fingerprinted Apache-2.0 (permissive) · 1cb732add54f1736 · report fixture: returned Swish · level recorded 2026-09-22
TimeReductionLayer squeezeformer/model.py ran Apache-2.0 (permissive) · ab409607eeb094c4 · report fixture: returned TimeReductionLayer · level recorded 2026-09-22
Transpose in 9 papers squeezeformer/model.py ran fingerprinted Apache-2.0 (permissive) · de0041a83f28b21f · report fixture: returned Transpose · level recorded 2026-09-22
recover_resolution squeezeformer/model.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · fe25270cda3bacd8 · report fixture: signature inputs:Tensor/3/float · returned Tensor · level recorded 2026-09-22
MultiHeadedSelfAttentionModule squeezeformer/model.py unverified Apache-2.0 (permissive) · 85cd1ee826586a47 · report
Squeezeformer squeezeformer/model.py unverified Apache-2.0 (permissive) · 4da7516805a0404e · report
SqueezeformerBlock squeezeformer/model.py unverified Apache-2.0 (permissive) · 96410b26ea1bc10d · report
SqueezeformerEncoder squeezeformer/model.py unverified Apache-2.0 (permissive) · b2b5c7140fffa95d · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

3 assemblies on this paper: 3 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

3 sibling assemblies for Squeezeformer (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Squeezeformer (graph method: Squeezeformer). 18 bricks; individually on a fixture: 14 ran (of which 12 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. Squeezeformer entry · level 0 · unverified · 4da7516805a0404e
  2. SqueezeformerEncoder model · level 0 · unverified · b2b5c7140fffa95d
  3. DepthwiseConv2dSubsampling helper · level 1 · ran · 98a41dcbe0265925
  4. ResidualConnectionModule helper · level 1 · ran · aabca50483e3eae8
  5. SqueezeformerBlock model · level 0 · unverified · 96410b26ea1bc10d
  6. TimeReductionLayer model · level 1 · ran · ab409607eeb094c4
  7. recover_resolution helper · level 1 · ran · our draft was wrong · fe25270cda3bacd8
  8. DepthwiseConv2d helper · level 1 · ran · 86a8c140aa415e79
  9. ConvModule model · level 1 · ran · a6e14d81239e64ad
  10. FeedForwardModule model · level 1 · ran · 425b3f6eba4ae90e
  11. MultiHeadedSelfAttentionModule helper · level 0 · unverified · 85cd1ee826586a47
  12. Swish helper · level 1 · ran · 1cb732add54f1736
  13. DepthwiseConv1d helper · level 1 · ran · f8183fb916aefe76
  14. GLU helper · level 1 · ran · 46a7b3e64f094c31
  15. PointwiseConv1d helper · level 1 · ran · 8acc35cd7fb4aa6f
  16. Transpose helper · level 1 · ran · de0041a83f28b21f
  17. RelPositionalEncoding helper · level 1 · ran · f528d6540416bf8b
  18. RelativeMultiHeadAttention helper · level 1 · ran · 0d0315fcda610792

harvested Assembly for Squeezeformer (graph method: Squeezeformer). 15 bricks; individually on a fixture: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 10 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. ConformerEncoder entry · level 0 · unverified · f8af5a4e29b98e92
  2. ConformerBlock model · level 0 · unverified · c6798a57b40a97c6
  3. Conv2dSubsampling helper · level 0 · unverified · 0c18f86c9e0954b7
  4. PositionalEncoding helper · level 1 · ran · 8008c17ae543b9f8
  5. TimeReductionLayer helper · level 0 · unverified · 480648d1fbf231db
  6. ConvFFModule model · level 0 · unverified · 664e0546c35af991
  7. ConvModule model · level 0 · unverified · c6e0c0f0739513ec
  8. FFModule model · level 1 · ran · 29208c0089f89df9
  9. IdentityLayer helper · level 1 · ran · 4d0d22b07f8f313e
  10. MHSAFFModule model · level 0 · unverified · 3065ed179d177d1a
  11. MHSAModule model · level 0 · unverified · 556aca63592449fd
  12. shape_list helper · level 1 · ran · our draft was wrong · f1f0e7aba532e2f6
  13. GLU helper · level 1 · ran · 35285e7d41e8ab62
  14. MultiHeadAttention helper · level 0 · unverified · 7d7f63e024f22b4b
  15. RelPositionMultiHeadAttention helper · level 0 · unverified · 9350e6b3773fb284

harvested Assembly for Squeezeformer (graph method: Squeezeformer). 16 bricks; individually on a fixture: 12 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 3 siblings

  1. SqueezeformerEncoder entry · level 0 · unverified · 74f45b8eb78d354f
  2. ConformerPreNet helper · level 0 · unverified · 77f464c34cf1fc3a
  3. SqueezeformerBlock helper · level 0 · unverified · 136716ec91fc7e49
  4. calc_data_len helper · level 1 · ran · honoured contract · 860c95bcb91d9a54
  5. get_mask_from_lens helper · level 1 · ran · our draft was wrong · c9d2ed75de0f8e2e
  6. Conv1DLayers helper · level 1 · ran · 4a641fcb9f0a3313
  7. AddAndNorm helper · level 1 · ran · 17b0a7a2343ae12d
  8. SqueezeformerConvModule helper · level 1 · ran · 4dea36bac3f898c4
  9. SqueezeformerFeedForward helper · level 1 · ran · e2853e88fa2a4612
  10. SqueezeformerRelativeMHSA helper · level 0 · unverified · 23a9c915a6989d9a
  11. ConformerConvModule helper · level 1 · ran · ac9fc9672ce1f61d
  12. Scaling1d helper · level 1 · ran · 8c469852dc208010
  13. ConformerFeedForward helper · level 1 · ran · e8be6862875569ea
  14. MultiHeadAtt helper · level 1 · ran · e3dbd3a46a9bf6f7
  15. add_pos_enc helper · level 1 · ran · honoured contract · cfc657760fd4bb85
  16. get_positional_encoding helper · level 1 · ran · honoured contract · 686e499ae419b99c

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2206.00888") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Speech Recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Recognition LibriSpeech test-clean Squeezeformer (L) Word Error Rate (WER) 2.47 #38 of 64 Archive leaderboard report
Speech Recognition LibriSpeech test-other Squeezeformer (L) Word Error Rate (WER) 5.97 #38 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

Concatenated Skip ConnectionConvolutionLayer NormalizationLinear LayerMax PoolingReLUSoftmaxU-Net

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