SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.06028 · 2020

TERA: Self-Supervised Learning of Transformer Encoder Representation for Speech

arXiv · PDF · Open in the Atlas

Code that ran

11 of 14 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 5 failures of our instrument; 3 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2007.06028.json · badge: /api/ran/2007.06028.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
Pandade1997/tera_asvproof community (classifier label)
classifier basis not recorded on the link
7 ran · 1 no recorded run
Vernacular-ai/Multimodal-Slu community (classifier label)
classifier basis not recorded on the link
4 ran · 2 no recorded run
FunctionStatusWhere it lives
swish Ran
ran; our contract draft was wrong, not the code
Pandade1997/tera_asvproof/transformer/model.py
code served (permissive licence) · get_code("0f786c407fb1ee4c")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-04T22:33:52Z). If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function swish takes a single parameter 'x' which is used with:
# - torch.sigmoid(x) - requires a tensor
# - x * ... - requires a tensor that supports multiplication
# So x should be a tensor

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
compare Ran
ran on a synthesized input
Vernacular-ai/Multimodal-Slu/utility/get_best_score.py
code served (permissive licence) · get_code("017e86bd00a963bc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function compares two values using the large_or_small operator
# a and b should be comparable (numbers, tensors, etc.)
# large_or_small should be '+' or '-'

a = torch.tensor(5.0)
b = torch.tensor(3.0)
large_or_small = '+'

ARGS = {
    'a': a,
    'b': b,
    'large_or_small': large_or_small
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
gelu Ran
ran, honoured the contract we drafted
Pandade1997/tera_asvproof/transformer/model.py
pointer only (licence: MIT); the stored copy get_code serves, from tanmay1618/text_to_sql_bert, is pointer only (licence: NONE) · get_code("fdc64f4c72036ae4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from tanmay1618/text_to_sql_bert, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_cosine_schedule_with_warmup Ran
ran on a synthesized input
Vernacular-ai/Multimodal-Slu/schedulers.py
code served (permissive licence) · get_code("28936c41e64792bd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned LambdaLR. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch.optim import SGD

torch.manual_seed(0)

# Create a simple model with parameters
model = nn.Linear(4, 3)

# Create an optimizer
optimizer = SGD(model.parameters(), lr=0.1)

# Define arguments for the function
ARGS = {
    "optimizer": optimizer,
    "num_warmup_steps": 100,
    "num_training_steps": 1000,
    "num_cycles": 0.5,
    "last_epoch": -1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type LambdaLR — a LambdaLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a LambdaLR output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_cosine_with_hard_restarts_schedule_with_warmup Ran
ran on a synthesized input
Vernacular-ai/Multimodal-Slu/schedulers.py
code served (permissive licence) · get_code("a9a4422e7a95a010")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned LambdaLR. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch.optim import SGD

torch.manual_seed(0)

# Create a minimal model
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

# Instantiate model and optimizer
model = MinimalModel()
optimizer = SGD(model.parameters(), lr=0.1)

# Define arguments
ARGS = {
    "optimizer": optimizer,
    "num_warmup_steps": 100,
    "num_training_steps": 1000,
    "num_cycles": 2,
    "last_epoch": -1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type LambdaLR — a LambdaLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a LambdaLR output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_grouped_parameters Ran
ran on a synthesized input
Vernacular-ai/Multimodal-Slu/optimizers.py
code served (permissive licence) · get_code("cecd81dc06f43149")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create small nn.Module instances with named parameters
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
        self.layer_norm = nn.LayerNorm(3)
    
    def forward(self, x):
        return self.layer_norm(self.linear(x))

class AnotherModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.embedding = nn.Embedding(10, 4)
        self.dense = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.dense(self.embedding(x))

# Instantiate models
model1 = SimpleModel()
model2 = AnotherModel()

# Build model_params - a list of nn.Module instances
model_params = [model1, model2]

# Define ARGS dictionary
ARGS = {
    'model_params': model_params
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
gumbel_softmax Ran
ran; our contract draft was wrong, not the code
Pandade1997/tera_asvproof/transformer/model_quantize.py
code served (permissive licence) · get_code("24b061a6d0af9f9e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn.functional as F
from torch.autograd import Variable

torch.manual_seed(0)

# The function gumbel_softmax expects:
# - logits: a tensor of shape (*, n_class)
# - temperature: a scalar temperature value

# Create a small logits tensor with shape (2, 4) - batch_size=2, n_class=4
logits = torch.randn(2, 4)

# Create a temperature scalar
temperature = 0.5

ARGS = {
    "logits": logits,
    "temperature": temperature,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
gumbel_softmax_sample Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Pandade1997/tera_asvproof/transformer/model_quantize.py
pointer only (licence: MIT); the stored copy get_code serves, from iamyuanchung/VQ-APC, is pointer only (licence: NONE) · get_code("4650349b06b3d90f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from iamyuanchung/VQ-APC, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_model Ran
ran on a synthesized input
Pandade1997/tera_asvproof/transformer/model_dual.py
code served (permissive licence) · get_code("8b8e06d04c4faf7f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned SimpleTransformer. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleTransformer(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        return self.linear2(self.linear1(x))

# Create a transformer model
transformer_model = SimpleTransformer()

# Create a state dict with some parameters
# Include 'gamma' and 'beta' keys to test the renaming logic
state_dict = {
    'linear1.weight': torch.randn(3, 4),
    'linear1.bias': torch.randn(3),
    'linear2.weight': torch.randn(2, 3),
    'linear2.bias': torch.randn(2),
    'layer_norm.gamma': torch.randn(4),
    'layer_norm.beta': torch.randn(4),
}

ARGS = {
    'transformer_model': transformer_model,
    'state_dict': state_dict,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SimpleTransformer — a SimpleTransformer output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleTransformer output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
prune_linear_layer Ran
ran; our contract draft was wrong, not the code
Pandade1997/tera_asvproof/transformer/model.py
code served (permissive licence) · get_code("d30d6c3098df2c42")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Linear. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from torch import nn

torch.manual_seed(0)

# Create a linear layer to prune
# Using distinct sizes: input_features=3, output_features=4
layer = nn.Linear(in_features=3, out_features=4, bias=True)

# Create an index tensor to select which output features to keep
# Selecting 2 out of 4 output features
index = torch.tensor([0, 2], dtype=torch.long)

# dim=0 means we're pruning along the output dimension (rows of weight matrix)
dim = 0

ARGS = {
    "layer": layer,
    "index": index,
    "dim": dim,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Linear — a Linear output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Linear output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
sample_gumbel Ran
ran; our contract draft was wrong, not the code
Pandade1997/tera_asvproof/transformer/model_quantize.py
pointer only (licence: MIT); the stored copy get_code serves, from iamyuanchung/VQ-APC, is pointer only (licence: NONE) · get_code("cec667e77fe8894d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from iamyuanchung/VQ-APC, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_optimizer Not yet run
unverified (harvested, no recorded run)
Vernacular-ai/Multimodal-Slu/optimizers.py
code served (permissive licence) · get_code("15a4c381f6af50fb")
How it ran

no invocation recorded.

get_scheduler Not yet run
unverified (harvested, no recorded run)
Vernacular-ai/Multimodal-Slu/schedulers.py
code served (permissive licence) · get_code("880bf296f53cdebf")
How it ran

no invocation recorded.

test_transformer Not yet run
unverified (harvested, no recorded run)
Pandade1997/tera_asvproof/run_upstream.py
code served (permissive licence) · get_code("78826dfaa4289758")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 13 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 4 the same month, 10 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
swish 206 other papers
42 dated earlier, 1 the same month, 162 later, 1 same year, order not determinable · 44 only as an assembly brick
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1805.12114 (2018-05)
Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
1809.01357 (2018-09)
Zero Shot Learning for Code Education: Rubric Sampling with Deep Learning Inference
and 203 more not named here
gelu 198 other papers
55 dated earlier, 1 the same month, 140 later, 2 same year, order not determinable · 32 only as an assembly brick
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.10125 (2019-01)
Glyce: Glyph-vectors for Chinese Character Representations
and 195 more not named here
prune_linear_layer 29 other papers
11 dated earlier, 1 the same month, 16 later, 1 same year, order not determinable
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1906.08230 (2019-06)
Evaluating Protein Transfer Learning with TAPE
1906.08237 (2019-06)
XLNet: Generalized Autoregressive Pretraining for Language Understanding
and 26 more not named here
gumbel_softmax_sample 7 other papers
3 dated earlier, 1 the same month, 3 later
1904.03240 (2019-04)
An Unsupervised Autoregressive Model for Speech Representation Learning
1906.08862 (2019-06)
Neural Stored-program Memory
2005.08392 (2020-05)
Vector-Quantized Autoregressive Predictive Coding
and 4 more not named here
get_cosine_schedule_with_warmup 5 other papers
5 later
2010.14150 (2020-10)
FragmentVC: Any-to-Any Voice Conversion by End-to-End Extracting and Fusing Fine-Grained Voice Fragments With Attention
2105.01051 (2021-05)
SUPERB: Speech processing Universal PERformance Benchmark
2208.11761 (2022-08)
IndicSUPERB: A Speech Processing Universal Performance Benchmark for Indian languages
and 2 more not named here
compare 3 other papers
3 later
2105.01051 (2021-05)
SUPERB: Speech processing Universal PERformance Benchmark
2208.11761 (2022-08)
IndicSUPERB: A Speech Processing Universal Performance Benchmark for Indian languages
2210.07185 (2022-10)
On the Utility of Self-supervised Models for Prosody-related Tasks
get_cosine_with_hard_restarts_schedule_with_warmup 3 other papers
3 later
2105.01051 (2021-05)
SUPERB: Speech processing Universal PERformance Benchmark
2208.11761 (2022-08)
IndicSUPERB: A Speech Processing Universal Performance Benchmark for Indian languages
2210.07185 (2022-10)
On the Utility of Self-supervised Models for Prosody-related Tasks
get_grouped_parameters 3 other papers
3 later
2105.01051 (2021-05)
SUPERB: Speech processing Universal PERformance Benchmark
2208.11761 (2022-08)
IndicSUPERB: A Speech Processing Universal Performance Benchmark for Indian languages
2210.07185 (2022-10)
On the Utility of Self-supervised Models for Prosody-related Tasks
get_optimizer 3 other papers
3 later
2105.01051 (2021-05)
SUPERB: Speech processing Universal PERformance Benchmark
2208.11761 (2022-08)
IndicSUPERB: A Speech Processing Universal Performance Benchmark for Indian languages
2210.07185 (2022-10)
On the Utility of Self-supervised Models for Prosody-related Tasks
get_scheduler 3 other papers
3 later
2105.01051 (2021-05)
SUPERB: Speech processing Universal PERformance Benchmark
2208.11761 (2022-08)
IndicSUPERB: A Speech Processing Universal Performance Benchmark for Indian languages
2210.07185 (2022-10)
On the Utility of Self-supervised Models for Prosody-related Tasks
sample_gumbel 3 other papers
3 dated earlier
1904.03240 (2019-04)
An Unsupervised Autoregressive Model for Speech Representation Learning
1906.08862 (2019-06)
Neural Stored-program Memory
2005.08392 (2020-05)
Vector-Quantized Autoregressive Predictive Coding
gumbel_softmax 2 other papers
2 dated earlier
1904.03240 (2019-04)
An Unsupervised Autoregressive Model for Speech Representation Learning
2005.08392 (2020-05)
Vector-Quantized Autoregressive Predictive Coding
test_transformer 1 other paper
1 dated earlier
1910.12638 (2019-10)
Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We introduce a self-supervised speech pre-training method called TERA, which stands for Transformer Encoder Representations from Alteration. Recent approaches often learn by using a single auxiliary task like contrastive prediction, autoregressive prediction, or masked reconstruction. Unlike previous methods, we use alteration along three orthogonal axes to pre-train Transformer Encoders on a large amount of unlabeled speech. The model learns through the reconstruction of acoustic frames from their altered counterpart, where we use a stochastic policy to alter along various dimensions: time, frequency, and magnitude. TERA can be used for speech representations extraction or fine-tuning with downstream models. We evaluate TERA on several downstream tasks, including phoneme classification, keyword spotting, speaker recognition, and speech recognition. We present a large-scale comparison of various self-supervised models. TERA achieves strong performance in the comparison by improving upon surface features and outperforming previous models. In our experiments, we study the effect of applying different alteration techniques, pre-training on more data, and pre-training on various features. We analyze different model sizes and find that smaller models are strong representation learners than larger models, while larger models are more effective for downstream fine-tuning than smaller models. Furthermore, we show the proposed method is transferable to downstream datasets not used in pre-training.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2007.06028")
get_code_for_paper("2007.06028")
have("2007.06028")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2007.06028.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2007.06028.svg)](https://syntology.ai/paper/2007.06028)

Connect an agent — have() is free.