In information technology, security has played a crucial role in protecting data. One of the methods that can be used is by hiding the data in a cover file, such as audio. However, the use of this carrier has several limitations, namely the capacity of the secret message and the quality of the stego data, which should not be easily recognized at the time of delivery. Therefore, in this research, we focus on overcoming these problems by exploring the modified high-order Lagrange interpolation to the audio samples equation. It is to produce a new signal as a space to embed relatively big confidential data. Furthermore, the Reduced Difference Expansion and smoothing algorithms are considered to maintain the quality of the stego audio. The evaluation is carried out on ten different sizes of payloads and 15 audio cover data. The results show that the use of the interpolation method can increase capacity.