This work presents a privacy analysis of iPhone 6 and iPhone 11 using acoustic side-channels, an under explored area for iPhone devices, particularly due to their complex System-on-Chip (SoC) Central Processing Units (CPUs). The study involves three key privacy attacks. First, a Side-Channel Based Disassembly (SCBD) is implemented using acoustic emissions from iPhone SoCs (Apple A8 and A13). By leveraging Power Spectral Density (PSD) of Sound Pressure Levels (SPL) and employing t-Distributed Stochastic Neighbor Embedding (t-SNE) for dimensionality reduction, the method achieves 100\% accuracy in template-based instruction classification. Second, an analysis of phone call recordings, featuring human speech, music, silence, and multiple running applications, successfully identifies the device model and running application with 100% and 90.9% accuracy, respectively. Third, the study explores using background noise from recorded calls to detect physical location changes between different calls. The research explores the feasibility of passive acoustic SCBD attacks on advanced systems like iPhones, despite challenges posed by background noise and concurrent CPU activities. The methodology involves profiling applications and instructions' template codes in noisy environments, isolating template codes using application-specific features. Testing assumes the template code is running on the device (e.g., via malware) and achieves 99.5% accuracy in instruction template classification from a phone call audio recording. Potential countermeasures and limitations of the proposed techniques are also discussed. Overall, this work underscores the significance of acoustic side-channels for security in SoCs and highlights the need for further research in this area.