Conformer + Wav2vec 2.0 + Noisy Student

Closed weights Google,Google Research,Google Brain 1B parameters October 2020

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
Google,Google Research,Google Brain
Organisation type
Industry,Industry,Industry
Country
United States of America
Published
20 October 2020
Authors
Yu Zhang, James Qin, Daniel S. Park, Wei Han, Chung-Cheng Chiu, Ruoming Pang, Quoc V. Le, Yonghui Wu

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Speech
Task
Speech recognition (ASR)

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
1B

1B for XXL model

Training data
tokens

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

Training compute
7.6 × 10²¹ FLOP

"We train with global batch size 2048 on 256/512 Google TPU V3 cores for 3-4 days for the XL/XXL models respectively... We fine-tune the pre-trained checkpoints (400k steps) with global batch size 1024/512 on 256/512 Google TPU v3 cores for 1-3 days for the XL/XXL models" TPU v3 chips are 123 teraflop/s. 2 chips per core 512 cores * 7 days * 24 * 3600 * 123 tflops * (1 chip/2 cores) * 0.4 (assumed utilization) = 7.6e21

How it was established
Hardware

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
Google TPU v3
Chips used
256
Wall-clock time
168 hours (7 days)

7 days

Power draw
234.3 kW
Compute cost
$9,450

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
Unreleased
Training code
Unreleased

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Why it is tracked
SOTA improvement

"By doing so, we are able to achieve word-error-rates (WERs) 1.4%/2.6% on the LibriSpeech test/test-other sets against the current state-of-the-art WERs 1.7%/3.3%."

Record confidence
Confident
Citations
332

Sources

Where this record came from and when it was last checked.

Reference
Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition
Last updated
25 May 2026

What the numbers mean

Where it came from

Conformer + Wav2vec 2.0 + Noisy Student was published by Google,Google Research,Google Brain, in the country recorded as United States of America, during October 2020. The publishing organisation is categorised as industry,Industry,Industry.

It works in the domain of Speech, and is recorded as performing the task of speech recognition (ASR).

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

Training and provenance

The training run consumed about 7.6 × 10²¹ FLOP, on hardware recorded as Google TPU v3. That figure measures what producing the model cost, and has no bearing on how fast it answers.

The reason it appears in this catalogue at all: sOTA improvement.

Answers

Conformer + Wav2vec 2.0 + Noisy Student — common questions

01

Conformer + Wav2vec 2.0 + Noisy Student— what is it used for?

It works in the domain of Speech, and is recorded as handling the task of speech recognition (ASR). These are the areas it was designed around; they describe intent rather than a hard boundary.

02

Conformer + Wav2vec 2.0 + Noisy Student— how much compute was used to train it?

Training consumed around 7.6 × 10²¹ FLOP, on hardware recorded as Google TPU v3. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.

03

Conformer + Wav2vec 2.0 + Noisy Student— what GPU do I need to run it?

None. This is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

04

Conformer + Wav2vec 2.0 + Noisy Student— is it open source?

No. Its weights have not been published, so it exists only as a service controlled by its owner.

05

Conformer + Wav2vec 2.0 + Noisy Student— how many parameters does it have?

It has a parameter count of 1B. 1B for XXL model. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

06

Conformer + Wav2vec 2.0 + Noisy Student— who created it?

It was published by Google,Google Research,Google Brain, based in United States of America, an organisation categorised as industry,Industry,Industry.

07

Conformer + Wav2vec 2.0 + Noisy Student— when was it released?

It was published in October 2020. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

Source

Original publication

Record last updated 25 May 2026

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.