Static word embedding model based on the Global Vectors architecture (Pennington et al., 2014). The embeddings provide real-valued vector representations for Tshivenḓa text.