Japanese Writing Support System with Fast Grammatical Error Correction

There are several problems in applying grammatical error correction (GEC) to a writing support system. One of them is the handling sentences middle input. Till date, performance GEC for incomplete not well-known. Hence, we analyze model sentences. Another problem speed. When speed slow, usability system limited, and user experience degraded. Therefore, this study, also focus on non-autoregressive (NAR) model, which widely studied fast decoding method. We perform Japanese with traditional autoregressive recent NAR models their accuracy Furthermore, construct function. Specifically, trained embedded back-end confirm system’s effectiveness by both objective subjective evaluations.