RNN-WER

Closed weights DeepMind,University of Toronto 26.5M parameters June 2014

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
DeepMind,University of Toronto
Organisation type
Industry,Academia
Country
United Kingdom of Great Britain and Northern Ireland, Canada
Published
22 June 2014
Authors
Alex Graves, Navdeep Jaitly

What it does

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

Domain
Speech
Task
Speech recognition (ASR)
Approach
Supervised

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
26.5M

"The network had five levels of bidirectional LSTM hidden layers, with 500 cells in each layer, giving a total of ∼ 26.5M weights."

Training data
tokens

dataset is 81 hours At 228 wpm (https://docs.google.com/document/d/1G3vvQkn4x_W71MKg0GmHVtzfd9m0y3_Ofcoew0v902Q/edit) that's 81*228*60 = 1,108,080 another source says WSJ contains 37k sentences, so this would be ~30 words per sentence which seems high but roughly right: https://www.arxiv-vanity.com/papers/1903.00216/

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
Highly cited,SOTA improvement

"Finally, by combining the new model with a baseline, we have achieved state-of-the-art accuracy on the Wall Street Journal corpus for speaker independent recognition."

Record confidence
Likely
Citations
2,805

Sources

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

Reference
Towards End-To-End Speech Recognition with Recurrent Neural Networks
Last updated
28 November 2025

What the numbers mean

About this model

RNN-WER was published by DeepMind,University of Toronto, in United Kingdom of Great Britain and Northern Ireland, in June 2014. industry,Academia is the category the publisher falls under.

It works in Speech, and is recorded as doing speech recognition (ASR).

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

Training and provenance

The reason it appears in this catalogue at all is highly cited,SOTA improvement.

Answers

RNN-WER — common questions

01

What is RNN-WER used for?

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

02

What GPU do I need to run RNN-WER?

None. RNN-WER 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.

03

Is RNN-WER open source?

The licensing for RNN-WER was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

04

How many parameters does RNN-WER have?

RNN-WER has 26.5M parameters. "The network had five levels of bidirectional LSTM hidden layers, with 500 cells in each layer, giving a total of ∼ 26.5M weights.". 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.

05

Who created RNN-WER?

RNN-WER was published by DeepMind,University of Toronto, based in United Kingdom of Great Britain and Northern Ireland, categorised as industry,Academia.

06

When was RNN-WER released?

RNN-WER was published in June 2014. 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 28 November 2025

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.