With the recent advent of Internet of Things (IoT), security poses a major critical concern of its practical implementation. It is imperative that the unique security challenges of IoT are resolved in order to realize the full potential of IoT systems. Current classical steganographic techniques might be impractical, complex, unscalable, and non-resilient to noise to be directly adaptable in IoT environments. In this paper, we address these issues by proposing a light-weight, noise-resilient, high payload audio steganography scheme capable of securing communication in IoT networks. The proposed technique is based on partial modification of well-selected phase frequencies leading to a smoother transition while preserving the phase continuity and the naturalness of the altered signal. The performance evaluation results show that the scheme encompasses high hiding capacity and low Bit Error Rate (BER), establishing it as a practical candidate for mass deployment of IoT devices.