In this modern era digital proof on case hoax not only in the form of chat, photos, and videos. Voice recording or audio too could used as goods digital evidence, like on cases certain recording sound have role important for showing involvement someone with something case for example hoax. Hoax or news lie who hasn't known the truth could more and more often appear especially on future presidential election. For identify something recording sound already experience editing or not yet need use audio forensic method that starts with stage acquisition, audio enhancement, decoding, and voice recognition. If recording goods evidence there is signal intersection means recording goods evidence already through the editing process or could said it as hoax because recording sound contain information that is not right. On stage voice recognition companion sound compared with some recording suspect for looking for know is friend goods evidence originated from the same person. compared on stage voice recognition is value pitch, value formant/bandwidth, value likelihood ratio (LR), and Graphical Distribution. This study aims to analyze the handling of hoax cases on recording sound with way compare with some suspect the suspected person is there on recording goods proof.