Melanoma is responsible for more than half of the deaths related to skin cancer in the last few decades. A dual-modality optical biopsy system with Raman spectroscopy and optical coherence tomography approach was built with the goal of achieving noninvasive skin measurement. To mimic melanoma and evaluate the effect of melanin on skin, models have been created by dissolving synthetic melanin in dimethyl sulfoxide and adding it to fresh skin samples. Compared to the untreated samples, morphological images showed that the imaging depth on melanin-treated skin has been increased from 250 μm to 350 μm due to the optical clearing effect of the DMSO solvent, and Raman analysis revealed that relative spectral intensities of melanin-treated samples were lower in the amide-I and CH2-deformation bands, and higher in the CH2-twist and C–C stretch bands. Using machine learning for skin type classification, an accuracy of 89% is achieved.