The aim of the study was to measure the performance reached by students (N = 138) when aurally identifying musical harmonic intervals (from m2 to P8) after having experienced a teaching innovation proposal for the Music Conservatories of Catalonia (Spain) based on observational methodology. Its design took into account several issues, which had been detected previously as being conflictive. The teaching innovation proposal was tested during an academic year in six Catalan Music conservatories, specifically in Music Theory classes in the first year of a 6-year professional course. A quasi-experimental pre-test/post-test design was used. The existence of a control group gave a reference point for the measured results before and after an intervention period. By the end of the teaching--learning process, it was established that the performance of the participants had improved, resulting in statistically significant differences in favour of the experimental group.