July 2018
Beginner to intermediate
312 pages
8h 31m
English
Statistical machine translation combines a translation model with a target language model to convert sentences from the source text in one language to sentences in the target language. This is illustrated in the following diagram. The translation model maps words and phrases from the source language to the target language. The language model captures statistics of how likely words follow a specific sequence in the target language. SMT, therefore, tries to maximize the probability of choosing a target sentence that is the translation of the source sentence. These statistical models are derived from a large corpus of source-to-target language translations:
Prior to SMT, machine translation involved defining ...
Read now
Unlock full access