Have you tried the H5P Speak the Words content type?
This is a simple example I created for my daughters as we started learning a little bit of Spanish.
Note that this H5P content type only works with Google Chrome browser, because it relies on Google code. I don’t usually create any content that isn’t truely cross browser, because a web browser that is made in the image of Google, or any other entity, is completely the opposite of what an open and free web should be !
The above said, here is an example of using the Speak the Words H5P content type to practice saying numbers 1 to 10 in Spanish – give it a try – it works quite well, although speech recognition is certainly not perfect.