The Web Salesman Analysis is a PHP tool that allow enterprises to test employees' skills. Once the employee records his voice and answers the question that is shown on the screen, a specialized Octave code runs in the background and gives him a table of results based on some criteria.
It is fundamental to understand that this is not a final product, but a demonstration that good mathematical tools can aid enterprises to build powerful data analysis that go beyond big data. A lot of tools can be attached to this software to make analysis even more accurate.
Below you can see the list of criteria that are used to analyze employees' voices
| Criteria | Description |
|---|---|
| Frequency | Men and women with low voices tend to be more powerful and attractive. High pitched voices are penalized |
| Power | Employees that are extremely loud or extremely silent are penalized |
| Frequency Variation | Checks whether the frequency of the voice varies. Monotone voices are penalized. |
| Power Variation | Checks whether the power of the voice varies. Monotone voices are penalized. |
| Semantic (Key Words) | The tool is able to understand which words are pronounced by the user, allowing to check whether the user is saying words that matter to the client. |
