Successful speech perception requires that listeners map the acoustic signal to linguistic categories. These mappings are not only probabilistic, but change depending on the situation. For example, one talker’s /p/ might be physically indistinguishable from another talker’s /b/ (cf. lack of invariance). We characterize the computational problem posed by such a subjectively non-stationary world and propose that the speech perception system overcomes this challenge by (1) recognizing previously encountered situations, (2) generalizing to other situations based on previous similar experience, and (3) adapting to novel situations. We formalize this proposal in the ideal adapter framework: (1) to (3) can be understood as inference under uncertainty about the appropriate generative model for the current talker, thereby facilitating robust speech perception despite the lack of invariance. We focus on two critical aspects of the ideal adapter. First, in situations that clearly deviate from previous experience, listeners need to adapt. We develop a distributional (belief-updating) learning model of incremental adaptation. The model provides a good fit against known and novel phonetic adaptation data, including perceptual recalibration and selective adaptation. Second, robust speech recognition requires listeners learn to represent the structured component of cross-situation variability in the speech signal. We discuss how these two aspects of the ideal adapter provide a unifying explanation for adaptation, talker-specificity, and generalization across talkers and groups of talkers (e.g., accents and dialects). The ideal adapter provides a guiding framework for future investigations into speech perception and adaptation, and more broadly language comprehension.