Flexible strain sensors have been widely researched in fields such as smart wearables, human health monitoring, and biomedical applications. However, achieving a wide sensing range and high sensitivity of flexible strain sensors simultaneously remains a challenge, limiting their further applications. To address these issues, a cross-scale combinatorial bionic hierarchical design featuring microscale morphology combined with a macroscale base to balance the sensing range and sensitivity is presented. Inspired by the combination of serpentine and butterfly wing structures, this study employs threedimensional printing, prestretching, and mold transfer processes to construct a combinatorial bionic hierarchical flexible strain sensor (CBH-sensor) with serpentine-shaped inverted-V-groove/wrinkling− cracking structures. The CBH-sensor has a high wide sensing range of 150% and high sensitivity with a gauge factor of up to 2416.67. In addition, it demonstrates the application of the CBH-sensor array in sign language gesture recognition, successfully identifying nine different sign language gestures with an impressive accuracy of 100% with the assistance of machine learning. The CBH-sensor exhibits considerable promise for use in enabling unobstructed communication between individuals who use sign language and those who do not. Furthermore, it has wide-ranging possibilities for use in the field of gesture-driven interactions in human−computer interfaces.