The embodiments of the application disclose a fundamental frequency extraction method and device. A specific implementation of the method includes the following steps: extracting the candidate fundamental frequency points of each speech frame in a speech signal to be processed based on the acoustic characteristics of the speech signal to be processed; classifying the voice frames to unvoiced or voiced sound to get the sound category corresponding to each speech frame; and correcting the candidate fundamental frequency points based on the sound category corresponding to each speech frame and apreset fundamental frequency screening condition, and using a dynamic programming algorithm to determine the fundamental frequency sequence of the speech signal to be processed from the corrected candidate fundamental frequency points. The implementation improves the accuracy of fundamental frequency extraction.