In this experiment we evaluate phonetic training as a tool for language learning. Specifically, we take a group of native speakers (NS) of English (n = 24) currently enrolled in Arabic classes at American universities, and evaluate the effectiveness of a high variability phonetic training program (HVPT) to improve their perception of a difficult Arabic phoneme contrast. We find that a group of phonetic trainees showed significantly greater improvement in a minimal pair test of the Arabic /h/-/h/ contrast (F[subscript 1,22] = 8.89, p = 0.007, ?[subscript 2] = 0.288) compared to an untrained control group. We argue that such training in bottom-up decoding processes has been neglected in the literature of second language pedagogy and that phonetic training can be an important component of a foreign language curriculum, particularly in light of the low cost it imposes on both teachers and learners.