# What hardware do is needed for automated speech recognition (ASR) system?

### The following data was prepared for the year 2020.

:heavy\_check\_mark: Note that 1 vCPU can serve 4 concurrent ASR sessions over time.

:heavy\_check\_mark: RAM and HDD depend on the configuration of the virtual machines.

:heavy\_check\_mark: To calculate how many gigabytes are required for 1 server instance supporting N channels, you need to multiply the number of channels by 100 MB on top of obligatory the 4.5 GB.

 $$(4.5 GB + N \* 100 MB)$$

:heavy\_check\_mark: Each additional context model additionally requires at least 3GB RAM per 1 server instance.

![](https://lh4.googleusercontent.com/LaOTGQpmppn4ZtuCbS2MbuH28bAsGuADjh-rIFdLFj1LRxGAPbQY57jX8aBuszlys7vM04jZgcGWgvcoB6SpWhS-fb5x3_yvvTJ4yxkbLujkJoqZEXsG20iOwwZC-90uNp7nHOGW)

|                                              |             **10 channels**             |            **50 channels**            |                  **100 channels**                  | <p><strong>200</strong></p><p><strong>channels</strong></p> |
| -------------------------------------------- | :-------------------------------------: | :-----------------------------------: | :------------------------------------------------: | :---------------------------------------------------------: |
| **Virtual Machines**                         |                    1                    |                   1                   |                          2                         |                              4                              |
| **CPU Threads**                              |                  3vCPU                  |                12 vCPU                |            <p>2x12 = </p><p>24 vCPU</p>            |                 <p>4x12 =</p><p>48 vCPU</p>                 |
| **CPU points**                               |                 2000 CPU                |             2000 - 2300CPU            |                   2000 - 2300CPU                   |                         Min 2300CPU                         |
| **Recommended Hardware**                     | <p>Xeon Gold   6136</p><p>or better</p> | <p>Xeon Gold 6136</p><p>or better</p> | <p>Xeon</p><p>Gold </p><p>6136</p><p>or better</p> |                 2 x Xeon Gold 6136 or better                |
| **Basic RAM**                                |                   8 GB                  |                 16 GB                 |                  2 x 16 GB = 32 GB                 |               <p>4 x 16 GB = </p><p>64 GB</p>               |
| **RAM in case u use up to 2 context models** |                   16GB                  |                  24GB                 |                  2 x 24 GB = 48GB                  |   <p>4 x 24 GB =</p><p>96 GB</p><p>recommended 128 GB</p>   |
| **HDD**                                      |                  40 GB                  |                 60 GB                 |                 2 x 60 GB = 120 GB                 |                      4 x 60 GB = 240 GB                     |


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://help.inteliwise.com/how-to-connect-a-voicebot-with-a-phone-number/what-is-required-for-the-on-premise-implementation/what-hardware-do-is-needed-for-automated-speech-recognition-asr-system.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.