Character encoding in TTL file

Question

I have the following TTL file (exported from WebProtege) which includes food ingredients like the following:

:water rdf:type :FoodIngredient ;
       skos:prefLabel "Water" ;
       skos:altLabel "Acqua"@it ,
                     "Agua"@es ,
                     "Eau"@fr ,
                     "Wasser"@de ,
                     "água"@pt ,
                     "�ΕΡΟ"@el ,
                     "アイスキューブ"@ja ,
                     "æ°´"@ja ,
                     "æ°·"@ja .

I guess the character encoding of the file does not permit the correct rendering of the last 2 labels, so is there a way to retrieve the correct labels?

EDIT: After following teapot418's suggestion (see comments section), I did iconv -c -f utf8 -t latin1 and I now have the following output:

:water rdf:type :FoodIngredient ;
       skos:prefLabel "Water" ;
       skos:altLabel "Acqua"@it ,
                     "Agua"@es ,
                     "Eau"@fr ,
                     "Wasser"@de ,
                     "água"@pt ,
                     "�ΕΡΟ"@el ,
                     "アイスキューブ"@ja ,
                     "水"@ja ,
                     "氷"@ja .

Any help on resolving the encoding issue with the two erroneous labels would be highly appreciated!