A novel model-based biclustering approach for multivariate data is introduced exploiting a finite mixture of generalized latent trait models. The proposed model clusters units into distinct subsets, called components. Within each component, subsets of variables, called segments, are identified by specifying the linear predictor in terms of a row-stochastic vector. The continuous latent trait integrated into the model allows us to account for the residual dependence between multivariate outcomes from the same unit. We employ an EM algorithm for maximum likelihood estimation of model parameters, with Gauss-Hermite quadrature utilized to approximate multidimensional integrals where closed-form solutions are not available.