Emotion migration speech synthesis method and system

The invention relates to the technical field of speech synthesis, in particular to an emotion migration speech synthesis method and system. Comprising the steps of obtaining a text coding vector; obtaining an emotion style vector; a text-voice alignment sequence is obtained; and inputting the spokesman identity ID into a voice frame decoder, processing the text-voice alignment sequence, and decoding to obtain Mel sound spectrum features. According to the emotion information extraction module provided by the invention, spokesman information and emotion information in audio features can be completely decoupled, an emotion coding vector only contains the emotion information in the audio, and the similarity between the coding vector and the emotion information represented by the vector is improved; the emotion coding vector can be freely combined with the spokesman information, so that the task of migrating the emotion information to the non-emotion target spokesman from the audio data of the source spokesman is