Minimization of Sequential Transducers

Mehryar Mohri

Abstract: We present an algorithm for minimizing sequential transducers. This algorithm is shown to be efficient, since in the case of acyclic transducers it operates in O(¦E¦+¦V¦+(E¦−¦V¦+¦F¦).(¦P max ¦+1) steps, where E is the set of edges of the given transducer, V the set of its vertices, F the set of final states, and P max the longest of the greatest common prefixes of the output paths leaving each state of the transducer. It can be applied to a larger class of transducers which includes subsequential transducers.