called CoBaLT (Corpus-Based Lexicon Tool), developed to construct corpusbased computational lexica and to correct word-level annotations and transcription errors in corpora. The paper describes the tool as well as our experience in using it to annotate a reference corpus and compile a large lexicon of historical Slovene. The annotations used in our project are modern-day word form equivalent, lemma, part-of-speech tag and optional gloss. The CoBaLT interface is word form oriented and compact. It enables wildcard word searching and sorting according to several criteria, which makes the editing process flexible and efficient. The tool accepts preannotated corpora in TEI P5 format and is able to export the corpus and lexicon in TEI P5 as well. The tool is implemented using the LAMP architecture and is freely available for research purposes.