Abstract: In recent years, people have increasingly relied on public Internet platforms for the free release and access of information. At the same time, the proliferation of fake information in cyberspace has become increasingly severe, bringing arduous challenges to people's daily lives and social governance. Therefore, the effective identification of fake information in cyberspace has become an urgent problem to be solved. In recent years, with the development of deep neural network technology, it has also brought new development vitality to false information generation and detection technology, which has led to the rapid development of these two countermeasures. This paper mainly focuses on the detection and forensics technology of fake remarks on the Internet, and summarizes the research results related to fake remarks detection and forensics on the text and voice respectively. This paper mainly focuses on the forensic technologies of fake Internet remarks, and summarizes the related works from two aspects: text and voice. In terms of fake text forensics, this paper summarizes related researches from three perspectives: text content authenticity verification, text semantic feature analysis, and communication mode analysis. In terms of fake speech forensics, this paper first introduces the current development of speech synthesis, speech conversion extension and other technologies, and then summarizes the latest research works on speech authentication and forensics technology. We hope that this review can help researchers in relevant fields better understand the latest research progress in the field of fake Internet remarks forensics, so as to develop more advanced technology and promote the continuous progress in this field.