The protocol describes quantitative analysis of males’ phonotaxis induced by female calls. This protocol is used to demonstrate that the ultrasonic concave-eared torrent frogs Odorrana tormota inhabiting noise habitats, as a model of acoustic communicating animals, show hyperacute phonotaxis to the female’s courtship calls (1). The acoustic recording and playback systems used have several advantages over the more widely used call or song detection methods: 1) the systems allow continuously recording diverse types of calling with a sampling rate (96-480 kHz); 2) calls can be simultaneously analyzed and displayed using SELENA software; 3) recorded calls can be easily saved into electronic data files and chosen as stimuli, and 4) the stimuli are varied according to experimental need and stably presented through the playback device for auditory electrophysiological or ethological study. The method for quantitative analysis of phonotaxis includes that 1) the scaled traces of acoustic evoked phonotactic paths are based on the videotaped record of the experiment; 2) five parameters are measured: latency and pattern of antiphonal response, distance of a hop, azimuthal angle and index of straightness (2) relative to the broadcasting loudspeaker based on unambiguous phonotactic behaviors, and 3) all data are statistically tested.