Methods of imposing native speakers' prosody on non-native speakers' utterances

摘要

A methods for studying the prosody of the native speaker using voice synthesis technology is provided to double study effect by overlapping prosodic features, such as intonation or the duration of consonants and vowels, on the sentences of a non-native learner and then letting the learn hear the prosody of the native speaker as the own voice. A methods for studying the prosody of the native speaker using voice synthesis technology comprises steps of: making the sound of the native speaker and the non-native speaker into sound files including pauses; making labels for respective phonemes of the sound file one by one by an automatic phoneme recognition function or manual work; converting the duration of each phoneme correspondently to the phoneme duration of the native speaker or the non-native speaker by using a voice synthesis and conversion program; converting intonation patterns of the two sound files including the phonemes with the same duration, to be same as the intonation patterns of the native speaker by using the voice synthesis and conversion program; and converting the intensity curve of the two sound files including the phonemes with the same duration, to be same as the native intensity.