Subtle vibrations, such as sound and ambient noises, are common mechanical waves that can transmit energy and signals for modern technologies such as robotics and health management devices. However, soft electronics cannot accurately distinguish ultrasmall vibrations owing to their extremely small pressure, complex vibration waveforms, and high noise susceptibility. This study successfully recognizes signals from subtle vibrations using a highly flexible anisotropic conductive gel (ACG) based on biphasic liquid metals. The relationships between the anisotropic structure, subtle vibrations, and electrical performance are investigated using rheological–electrical experiments. The refined anisotropic design successfully realized low-cost flexible electronics with ultrahigh sensitivity (Gauge Factor: 12787), extremely low detection limit (strain: 0.01%), and excellent frequency recognition accuracy (>99%), significantly surpassing those of current flexible sensors. The ultrasensitive flexible electronics in this study are beneficial for diverse advanced technologies such as acoustic engineering, wearable electronics, and intelligent robotics.