A Corpus Building Method on Forensic Identification of Dialects (FIDS)

摘要

China is very extensive in area and is with a large population, which leads to many dialects. In such a case dialects with accents to some degree should be analyzed to ascertain speakers' identification in legal practice. Here, a corpus building method is proposed and demonstrated, which shows how to project a corpus for forensic identification of dialect speakers through an examle of Puyang city. The design ideas, the design plans, recording operation, and technologies of processing collect data in the course of projecting such a corpus are all introduced. The analysis shows that the proposed method can solve the problem of forensic identification of dialects.