Large coil arrays are widely used in clinical routine for cardiovascular imaging providing extended spatial coverage and enabling accelerated acquisition using parallel imaging approaches. This work investigates the use of large coil arrays in single-voxel cardiac spectroscopy for the detection of myocardial creatine and triglyceride content. For this purpose, a navigator-gated and cardiac-triggered point-resolved spectroscopy sequence was implemented, and data obtained in 11 healthy volunteers using 32- and 5-element coil arrays were compared. For combination of the individual coil element signals, four strategies were evaluated differing in the manner of estimation of the complex coil weights and the amount of additional information required for coil combination. In all volunteers, and with both the 32- and 5-channel coil arrays, triglyceride-to-water (0.44 ± 0.19% and 0.45 ± 0.17%) and total creatine-to-water (0.05 ± 0.02% and 0.05 ± 0.01%) contents were computed. The values were found to agree well, showing an intraclass correlation coefficient of 0.76 (p < 0.003). The results revealed a gain in signal-to-noise ratio of approximately 24% with the 32-channel coil relative to the 5-channel array. The findings may foster the integration of cardiac spectroscopy into clinical practice using large coil arrays, provided that appropriate reconstruction algorithms are implemented.