Detect the text language automatically using a bigram model, Support Vector Machines, and Artifical Neural Networks. The model is trained using the WiLI-2018 benchmark dataset, and the highest accuracy achieved on the test dataset is 99.7% with paragraph text.