This could just be an issue of minimal pairs in the speech of the individual recorded, but the audio files for cmn-zhu2 and cmn-zhu4 sound exceedingly close to cmn-bu2 and cmn-bu4. I compare the audio for cmn-zhu2 to zdic entry for 竹 - the latter has extremely clear zh sound.
This could just be an issue of minimal pairs in the speech of the individual recorded, but the audio files for cmn-zhu2 and cmn-zhu4 sound exceedingly close to cmn-bu2 and cmn-bu4. I compare the audio for cmn-zhu2 to zdic entry for 竹 - the latter has extremely clear zh sound.