MS-ensemble-speech-recognition

Closed weights Microsoft 3.2B parameters September 2016

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
Microsoft
Organisation type
Industry
Country
United States of America
Published
12 September 2016
Authors
Wayne Xiong, J. Droppo, Xuedong Huang, F. Seide, M. Seltzer, A. Stolcke, Dong Yu, G. Zweig

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
3.2B

Large ensemble CNNs: 85000000 + 38000000 + 65000000 = 188000000 = 188M LSTMs: 40 / 140 input - 6*512 layers - 9000 / 27000 output 4*(140+512)*512 + 5*4*(512+512)*512 + 4*(512+27000)*27000=2983117056 LM: 1000*1000 = 1000000 + Embedding Total: 1000000+2983117056+188000000=3172117056 (underestimate because it doesn't account for embeddings)

Training data
11,140,000,000 tokens

2000h of audio data + 85M words of text data

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

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

https://paperswithcode.com/sota/speech-recognition-on-switchboard-hub500

Record confidence
Likely

Sources

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

Reference
The microsoft 2016 conversational speech recognition system
Last updated
28 November 2025

What the numbers mean

Background

MS-ensemble-speech-recognition was published by Microsoft, in United States of America, in September 2016. The organisation is categorised as industry.

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

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

What went into building it

Around 11,140,000,000 tokens went into training it.

Its inclusion criterion is sOTA improvement.

Answers

MS-ensemble-speech-recognition — common questions

01

What is MS-ensemble-speech-recognition used for?

MS-ensemble-speech-recognition 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 MS-ensemble-speech-recognition?

None. MS-ensemble-speech-recognition 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 MS-ensemble-speech-recognition open source?

No. MS-ensemble-speech-recognition has not had its weights published, so it exists only as a service controlled by its owner.

04

How many parameters does MS-ensemble-speech-recognition have?

MS-ensemble-speech-recognition has 3.2B parameters. Large ensemble CNNs: 85000000 + 38000000 + 65000000 = 188000000 = 188M LSTMs: 40 / 140 input - 6*512 layers - 9000 / 27000 output 4*(140+512)*512 + 5*4*(512+512)*512 + 4*(512+27000)*27000=2983117056 LM: 1000*1000 = 1000000 + Embedding Total: 1000000+2983117056+188000000=3172117056 (underestimate because it doesn't account for embeddings). 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 MS-ensemble-speech-recognition?

MS-ensemble-speech-recognition was published by Microsoft, based in United States of America, categorised as industry.

06

When was MS-ensemble-speech-recognition released?

MS-ensemble-speech-recognition was published in September 2016. 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.