Split pit in peach (Prunus persica (L.) Batsch) fruit is an internal disorder that can have adverse effects on fruit quality and shelf stability. As it is difficult to distinguish split-pit fruit from normal fruit by appearance, a nondestructive detection method is desirable to determine accurately the timing of split-pit occurrence and remove unwanted fruit from trees. In our recent study, we used an acoustic vibration method for the nondestructive detection of split pit in harvested peach fruit. Here, we demonstrate that this method can be used to detect split pit in unpicked peach fruit. Time-course monitoring of growing fruit revealed that the ratio of the third (f₃) to the second (f₂) resonant frequency (f₃/f₂), which is an indicator of split pit in harvested fruit, abruptly increased in unpicked split-pit fruit from early June to early July. In contrast, the f₃/f₂ values of normal fruit remained low until harvest in late July. The increase in f₃/f₂ values of split-pit fruit was observed in early June (stage II of fruit growth) and/or from late June to early July (stage III of fruit growth), suggesting that, at least under the experimental conditions in the current study, pit splitting occurred at these two different timings. Split-pit fruit detection rate at harvest increased towards the end of the second split-pit occurrence and split pit could be predicted with high accuracy thereafter. These results collectively suggested that the timing of split-pit occurrence in unpicked peach fruit could be predicted accurately by measuring f₃/f₂ values and unwanted fruit with split pit could be distinguished from normal fruit on trees. We discuss the possible applications of the nondestructive acoustic vibration method in combination with fruit thinning and the future use of this method in research aiming to develop effective prevention methods or resistant cultivars with reduced split pit.