This dataset is the semi-raw data used to build the Sinitic Lexical Network, browsable at https://silene.magistry.fr/ It is a compilation of various sources of lexical description of different languages for which sinograms are a traditional script. These sources were released as Open Data by different third parties and are combined here for convenience in cross-lingual linguistic studies. This work is a graph-based extention of a work started with Fabienne Marc on Mandarin Chinese in the early 2000's. It has since then shifted to a multilingual perspective on the Chinese script, with a larger team of young researchers. For the sake of simplicity, the format is a single CSV file where each row describes the reading of a sinogram in a specific word in a specific language. Its design was dictated by our need for the work on the book «À L'Écoute des Sinogrammes» (ALES), but it can be converted to build the SiLeNe graph.