A timestamp-based audio and video processing method and a system thereof are provided. The method is adapted to an electronic device having an image capturing device and an audio input device and includes the following steps. First, an image signal and an audio signal are extracted within a same time interval by respectively using the image capturing device and the audio input device so as to generate video data and audio data, and the time interval is converted to timestamp data, where the video data includes a single video frame with a variable file size, the audio data includes multiple audio files with a fixed file size. Next, the video data, the audio data, and the timestamp data are integrated into transmission data.