Looks like that pronunciation is just from one dictionary - the FLTRP Chinese-Italian - so can be ignored. (at some point we need to come up with a 'voting system' so that a single variation won't necessarily be elevated to the top)
As for the audio recordings - I can hear how it kind of sounds that way (partly with the northern accent 'ji-uh-ng') but I wouldn't necessarily say it's wrong; that being said, if you prefer that audio very precisely match the readings in the dictionary I would generally recommend using TTS instead.