Traditional airborne microphones are at risk of failure due to their dependence on airborne media in complex acoustic environments (CAEs). Here, we report an anti-interference self-powered acoustic fabric (ASAF) that can shield the interfering factors in the process of sound production and propagation to serve as a precise and wearable sound receiver. The use of the soft and safe woven structure polyvinylidene fluoride (PVDF) as a vibration-sensitive layer enables the ASAF to record human speech at wide vibration frequencies (0–5000 Hz). A speech recognition system is established which can recognize 25 words related to extreme weather conditions, with more than 95.8% accuracy. This speech recognition is carried out in CAEs such as wearing masks, silent communication, windstorms, and rainstorms, with corresponding losses of accuracy less than 1.6%, 6.7%, 6.3%, and 5.8%, respectively. The ASAF is expected to facilitate outdoor rescuers, journalists, students, and other professionals working in CAEs.

传统机载麦克风由于在复杂声学环境 (CAE) 中依赖机载介质而存在故障风险。在这里，我们报告了一种抗干扰自供电声学织物（ASAF），它可以屏蔽声音产生和传播过程中的干扰因素，作为一种精确的可穿戴声音接收器。使用柔软安全的编织结构聚偏二氟乙烯 (PVDF) 作为振动敏感层，使 ASAF 能够在宽振动频率 (0–5000 Hz) 下记录人类语音。建立了语音识别系统，可识别25个与极端天气情况相关的词，准确率超过95.8%。这种语音识别在戴口罩、无声交流、暴风雨等CAE中进行，相应的准确率损失小于1.6%，分别为 6.7%、6.3% 和 5.8%。ASAF 有望为户外救援人员、记者、学生和其他在 CAE 工作的专业人员提供便利。