Title-casing of lowercase messages is currently automated through ucfirst calls in the code. There are two issues with the current implementation:

1. Special characters are not upper-cased
2. Title-casing is not always correct

For example:

1. French émérite fails to be title-cased (to "Émérite")
2. English Musical relationships type is not correctly title-cased (to "Musical Relationships"), whereas it would be correct into French ("Relation musicale")

It might be best to avoid title-casing automation. If necessary, (localized) quotation marks might be added instead.