Motor imagery (MI) is a well-known task utilized within brain-computer interface (BCI) systems. Various studies have investigated MI-BCI and classification models to optimize the system’s performance; however, the obstacle of participants' low-performance ability in BCI task remains. A standard solution to low BCI performance is a training protocol. This study focuses on an audio-based neurofeedback training protocol. Music and audio feedback have been used within various motor applications, such as rehabilitation and sports training. Audio feedback is also known to show activity in both the audio and motor cortex, which is why audio stimulus is commonly used in motor applications. The audio feedback for motor applications is mainly generated from kinematic sensors, where the novelty of this study is to utilize electroencephalography (EEG) for audio generated feedback. Using one of the main motor rehabilitation techniques of therapeutic instrumental music performance (TIMP) and rhythmic auditory feedback, the effectiveness of this audio-based neurofeedback is investigated for MI-BCI performance training.