Splines have been used for the solution of the considered problems. This allows getting a single model for the three tasks and combine flexibility of the model with ease of calculations. For filtering and segmentation of acoustic signals spline filters that are similar to Savitsky-Golay filters have been used. Various widths of spline fragments provides a possibility to have different smoothness and select fragments of varying detalization. Selection tool for the heart tones is time-frequency LSS analysis, where decomposition is based on scaled spline approximation with the method of least squares. To distinguish significant timefrequency components, selection of coefficients by Student’s t-test is used. To take into account the presence of signals in different frequency bands, original decomposition algorithm of reverse estimate of the residuals is used. As the result we have the set of parameters of timefrequency transform that characterize cardiac signals in details. This allows comparing the similarity of tones at different periods of observation, to form a generalized tone and use these parameters for classification of heart sounds.