Capitalized text with "notranslate" MS Translator Text

Hi there,

I've an issue with the "notranslate" annotation for HTML translation.

Some background: We've an icon font which uses the CSS class "my-icon". During translation the icon has to be excluded from translation to not alter the icon. Thus we add the CSS class "notranslate" like in the documentation.

Here the doc: https://video2.skills-academy.com/en-us/azure/cognitive-services/translator/reference/v3-0-translate#translate-content-with-markup-and-decide-whats-translated

The problem: For some target languages the text inside the "notranslate" container is capitalized and for some target languages it's not.

Minimal example:

curl -X POST -v "https://api-eur.cognitive.microsofttranslator.com/translate?api-version=3.0&from=de&to=en&textType=html" -H "Ocp-Apim-Subscription-Key: XXXXXX" -H "Content-Type: application/json; charset=UTF-8" -d "[{'Text':'<span class=\"my-icon notranslate\">m</span>'}]"

Response with upper case "M" instead of "m":

[
{
"translations": [
{
"text": "<span class=\"my-icon notranslate\">M</span>",
"to": "en"
}
]
}
]

When changing the target language to "fr" it remains lower case. Thus I guess it's not indented.

Does anyone has a clue if this is "correct" behavior or not?