The invention provides a method of evaluating the quality of an audiovisual sequence by:a) training, comprising allocating a subjective score NS.sub.i to each of N.sub.0 training sequences S.sub.i (where i=1, 2, . . . , N.sub.0) presenting degradations identified by a training vector MO.sub.1 which is given to each sequence S.sub.i in application of a first vectorizing method, in order to build up a database of N.sub.0 training vectors MO.sub.i with subjective scores NS.sub.1;b) classifying the N.sub.0 training vectors MO.sub.i into k classes of scores as a function of the subjective scores NS.sub.1 that have been allocated to them, so as to form k training sets EA (where j=1, 2, . . . , k) which have k significant training scores NSR.sub.j allocated thereto;c) for each audiovisual sequence to be evaluated, generating a vector MO using said first vectorization method; andd) allocating to the audiovisual sequence for evaluation the significant training score NSR.sub.j that corresponds to the closest training set EA.sub.j.