Title: Inference of the sound speed and related properties of neutron stars

Abstract: Information on the phase structure of strongly interacting matter at high baryon densities can be gained from observations of neutron stars and their detailed analysis. In the present work Bayesian inference methods are used to set constraints on the speed of sound in the interior of neutron stars, based on recent multi-messenger data in combination with limiting conditions from nuclear physics at low densities. Two general parametric representations are introduced for the sound speed $c_s$ in order to examine the independence with respect to choices for the parametrisation of Priors. Credible regions for neutron star properties are analysed, in particular with reference to the quest for possible phase transitions in cold dense matter. The evaluation of Bayes factors implies extreme evidence for a violation of the conformal bound, $c_s^2 \leq 1/3$, inside neutron stars. Given the presently existing data base, it can be concluded that the occurrence of a first-order phase transition in the core of even a two-solar-mass neutron star is unlikely, while a continuous crossover cannot be ruled out. At the same time it is pointed out that the discovery of a superheavy neutron star with a mass $ M \sim 2.3 - 2.4\, M_\odot$ would strengthen evidence for a phase change in the deep interior of the star.