The climate emergency compels us to rethink how we conduct machine listening research. Against the business-as-usual narrative, we must anticipate new demands, new constraints, and new risks. Yet, the current literature tends to oversimplify the problem by reducing it to mere GPU usage. In this talk I will present three personal perspectives on our duty as audio-focused scientists in a warming world with finite energy supply. First, I will stress the importance of documenting the role of sound in ecological processes such as migration, species assemblages, and ecosystem decline. Secondly, I will describe an early-stage prototype of eco-acoustic sensor which operates without wires nor batteries by harvesting solar energy and performing sound event detection on device. Thirdly, I will show that working at the interface between sound and climate requires new methods in applied statistics, a critical lens on Shannon's theory of communication, and an inspiration from the musical concept of "ecological listening“.