Multilanguage model [DA,EN,NO]

[mustass]

Bjørn has posted a link to a dataset with danish and norwegian speech. What if we merge LJ with that one and add language as a condition.

[panosapos]

Since Danish Dataset is sampled at 16 kHz, we also need to check what changes need to be done in the extracted spectrograms