Аннотация:In this article,we consider the solution of an inverse problem of Raman spectroscopy of water-ethanol solutions by artificial neural networks (NN). Since training of a NN requires a large dataset, which often cannot be collected by laboratory measurements, we propose the approach of generating synthetic patterns represented by continuous vectors. This was achieved by using the partial least squares (PLS) model as an additional embedding model of latent space generated by conditional variational autoencoder (cVAE). Random sampling procedure generates vectors in PLS score space, which allows assigning a vector in the cVAE latent space and an output vector of the inverse problem. The generated patterns have a shape similar to real experimental spectra, and they are used to train the NN on the multi-output regression task, along with the original patterns. Applying this approach results in significant improvement of the quality of the inverse problem solution on real out-of-sample patterns not used in the training process.