The rapid advancement of artificial intelligence and mobile communication technologies has brought unprecedented convenience, but also significant privacy and security concerns, particularly in real-time video communications, such as deepfake. Several researchers have investigated multiple security measures, with Covert Timing Channels (CTCs) emerging as a promising technique for secure information transmission, yet most of them suffer from low transmission efficiency, limiting their real-time responsiveness and network adaptability, making them inadequate for high-capacity covert information transmission. To address this limitation, this paper proposes a novel CTC scheme that modulates the Inter-Frame Delay (IFD) between audio and video frames during VoLTE(Voice over LTE) video calls to transmit covert information. Furthermore, to maintain the undetectability of the channel, a fitting algorithm is used to ensure that the cumulative distribution function (CDF) of the IFD of covert traffic aligns with that of legitimate traffic. The design also incorporates an embedding factor that allows communicating parties to dynamically adjust network parameters, balancing robustness, transmission efficiency, and undetectability. Experimental results demonstrate that, compared to existing methods, the proposed IFDCTC scheme achieves higher transmission efficiency, with a peak transmission rate of up to 30 bps, while maintaining acceptable bit error rates and strong undetectability.