The bulk update request #1523 is active.

change category chinese_audio (10) -> meta
change category english_audio (1314) -> meta
change category finnish_audio (0) -> meta
change category french_audio (20) -> meta
change category german_audio (3) -> meta
change category japanese_audio (142) -> meta
change category korean_audio (8) -> meta
change category spanish_audio (108) -> meta
create implication chinese_audio (10) -> sound (40335)
create implication english_audio (1314) -> sound (40335)
create implication finnish_audio (0) -> sound (40335)
create implication french_audio (20) -> sound (40335)
create implication german_audio (3) -> sound (40335)
create implication japanese_audio (142) -> sound (40335)
create implication korean_audio (8) -> sound (40335)
create implication spanish_audio (108) -> sound (40335)

Reason: Following the standard set by *_text tags.

EDIT: The bulk update request #1523 (forum #319820) has been approved by @gattonero2001.

I think I'd prefer *_speech to *_audio, since the latter can indicate an OST or something used for the given language region (which have no lyrics, or fandubbed lyrics), rather than the spoken words and dialect the tags are intended for.

EDIT:
See BUR #1026 in topic #29830

How common is it to refer to an OST with the term (language)_audio? I don't believe that would become a significant problem in e621.

Actually, I find it far more likely that (language)_speech would be used to describe text dialogue by people who don't read the wiki.

The bulk update request #1523 (forum #319820) has been approved by @gattonero2001.