Research on speech spoofing detection based on big data and machine learning

Spoof detection is increasingly important in practical life, and its importance in assisting customs inspections, public security detection, trial, telecom fraud and other aspects is self-evident. Speech spoof detection has attracted much attention because of its advantages such as strong concealment, non-contact and relatively objective data. However, in general, speech spoofing detection is still in the development stage. With the rise of big data and machine learning, more and more researchers combine spooling detection with machine learning. Based on the lie database, they use machine learning training model to greatly improve the efficiency and accuracy of spoofing detection. This paper introduces the development history of lie detection technology, the basic indicators of non-language degradation and the correlation between acoustic characteristics and lie detection, then introduces some speech lie detection databases and lie detection based on big data and machine learning, and finally gives the prospect of speech lie detection.