We introduce childPoeDE: the first corpus of German poetry for children comprising poems which are still read today and cover a wide range topics authors. ChildPoeDE contains poem texts both poem-level token-level metadata. Poem-level metadata includes information about anthologies authors, quantitative text features, rhyme lexical richness. Token-level covers word length, position frequency, p...

Sound-imitation words, a sound-related subset of onomatopoeia, are important for computer-human interaction and automatic tagging of sound archives. The main problem of automatic recognition of sound-imitation word is that the literal representation of such words is dependent on listeners and influenced by a particular cultural history. Based on our preliminary experiments of such dependency an...