This paper presents a new approach for grapheme-to-phoneme conversion based on morphology. With this approach, a high accuracy can be obtained, although not for all words a transcription is achieved. The principle of this approach is to automatically decompose an existing pronunciation lexicon into morpheme-similar units called pseudo-morphological units. The pronunciation of the pseudo-morphological units is also generated from the pronunciation lexicon. The pronunciation of unknown words is composed from the pronunciation of its morphemes. With this approach, it is possible to transcribe also abbreviations and words, where a differentiation between spelling and pronunciation-as-a-word is difficult for other approaches. Another advantage of this approach is that it is performed automatically. It can further be used for any language as long as a word based pronunciation lexicon is available.