An intelligent voice prompting method and an electronic device, which are applicable to the technical field of data processing. The method comprises: obtaining telephone numbers of target objects, and identifying the types of insurances that have been insured; inputting operation record data, the types of the insurances and insurance node dates into a factor sifting model corresponding to the type of the insurances, so as to obtain class prompting factors corresponding to the types of the insurances respectively; if the total quantity of the obtained factors is smaller than or equal to a preset quantity of factors, reading out voice data of each of the prompting factors for broadcasting; and if the total quantity of the factors is greater than the preset quantity of factors, sifting out a preset quantity of prompting factors with a highest priority, and then broadcasting the voice data of the sifted-out prompting factors. Excessive broadcasts are avoided while it is ensured that finally-outputted prompting factors are actually needed by target objects, thereby greatly improving the intelligence level of an intelligent voice system in telephone service processing.