After a stroke, about one third of the patients is faced with aphasia, a language disorder affecting speech comprehension and production. Persons with aphasia show very heterogeneous symptoms and vary in the degree to which each level of the speech processing chain (i.e. auditory, phonological or semantic processing) is affected. Although a precise diagnosis is vital for targeted intervention, behavioral testing can be very challenging due to potential co-morbid cognitive problems. The aim of this project is to apply a novel electroencephalography (EEG) analyzing technique in which the brain signal is decoded while patients listen to a 30-minute-long story. This EEG technique allows (1) to automatically detect language problems in persons with aphasia, (2) to determine in which speech processing stage the locus of the language disorder lies and cluster patients with similar profiles, and (3) to validate the EEG results with behavioral language tests. This novel technique of neural tracking of speech will result in an objective characterization of the language problems in stroke patients with aphasia, above and beyond behavioral measures.