Raman micro-spectroscopy (RMS) has been recently proposed for label-free phenotypic identification of human embryonic stem cells (hESC)-derived cardiomyocytes. However, the methods used for measuring the Raman spectra led to acquisition times of minutes per cell, which is prohibitive for rapid cell sorting applications. In this study we evaluated two measurement strategies that could reduce the measurement time by a factor of more than 100. We show that sampling individual cells with a laser beam focused to a line could eliminate the need of cell raster scanning and achieve high prediction accuracies (>95% specificity and >96% sensitivity) with acquisition times ∼5 seconds per cell. However, the use of commercially-available higher power lasers could potentially lead to sorting speeds of ∼10 cells per s. This would start to progress RMS to the field of cell sorting for applications such as enrichment and purification of hESC-derived cardiomyocytes.