Acoustico-articulatory inversion of unequal-length tube models through lattice inverse filtering

Constraints related to the Distinctive Regions and Modes (DRM) speech production model are incorporated in the framework of speech analysis by inverse filtering. It is shown that the analogy between Auto-Regressive modeling and acoustic models based on acoustic tubes is still respected when using tubes with unequal length elementary sections. This analogy can be exploited for the inversion of the synthesis process using a method related to Burg's estimation method. Experimental results show that an improvement over traditional methods is brought by the DRM-related constraints included in the parameter estimation scheme.