Mel spectrum prediction method and device, equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a Mel spectrum prediction method, device and equipment and a storage medium, and the method comprises the steps: inputting a to-be-predicted text sequence into a text encoder of a target acoustic module for feature extraction, and obtaining target text coding feature data; performing alignment position prediction on the target text coding feature data through an alignment position predictor of the target acoustic module to obtain target alignment position data; performing time alignment feature calculation according to the target text coding feature data and the target alignment position data through an alignment graph reconstructor of the target acoustic module to obtain a target time alignment feature value; and performing Mel spectrum calculation on the target time alignment feature value through a decoder of the target acoustic module to obtain target Mel spectrum data. The time length modeling is implicitly integrated in the target acoustic module by adopting an input-output feature alignment strategy, and an additional time length model is not needed. The invention further relates to a block chain technology.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, in particular to a prediction method, device, equipment and storage medium of a mel spectrum. Background technique [0002] Singing synthesis is a technology that converts information such as lyrics and music scores into singing audio. With the popularization of the mobile Internet and the continuous improvement of people's requirements for the quality of entertainment and life, singing synthesis technology has gradually emerged in the fields of video games, short video applications, and virtual singers. [0003] In the existing singing synthesis method, when predicting the mel spectrum, it is necessary to obtain the phoneme / note duration information through manual labeling or automatic machine labeling, and additionally train a duration model based on this, and may even need to carry out the duration predicted by the duration model. Post-processing leads to cumbersome ...

AI Technical Summary

Problems solved by technology

[0004] The main purpose of this application is to provide a prediction method, device, equipment and storage medium of a Mel spectrum, aiming at solving the singing synthesis method of the prior art, when performing Mel spectrum prediction, an additional duration model needs to be trained, resulting in a model Technical problems with cumbersome and complicated training steps

Images