Unlock instant, AI-driven research and patent intelligence for your innovation.

Method for training acoustic conversion model, terminal and storage medium

A conversion model and acoustic technology, applied in the Internet field, can solve problems such as phoneme information error and inaccurate acoustic conversion model, and achieve the effect of suppressing influence and accurate acoustic conversion model

Pending Publication Date: 2021-06-18
TENCENT MUSIC ENTERTAINMENT TECH SHENZHEN CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the actual process, there may be certain errors in the phoneme information obtained based on the forced alignment model, which will lead to inaccurate acoustic conversion models after training.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method for training acoustic conversion model, terminal and storage medium
  • Method for training acoustic conversion model, terminal and storage medium
  • Method for training acoustic conversion model, terminal and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0095] In order to make the purpose, technical solution and advantages of the present application clearer, the implementation manners of the present application will be further described in detail below in conjunction with the accompanying drawings.

[0096] figure 1 It is a schematic diagram of an implementation environment of a method for training an acoustic conversion model provided in an embodiment of the present application. see figure 1 , the implementation environment includes: a terminal 101 and a server 102.

[0097] The terminal 101 can be a smart phone, a game console, a desktop computer, a tablet computer, an MP3 (Moving Picture Experts Group Audio Layer III, moving picture expert compression standard audio layer 3) player, an MP4 (Moving Picture Experts Group Audio Layer IV, moving picture expert compression Standard Audio Level 4) At least one of a player and a laptop computer. The terminal 101 is connected to the server 102 through a wired network or a wirel...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a method for training an acoustic conversion model, a terminal and a storage medium, and belongs to the technical field of Internet. The method comprises the following steps: acquiring phoneme information and pitch information corresponding to each audio frame in a sample song audio of a target object, and acquiring reference spectrum feature information corresponding to each audio frame; inputting the phoneme information and the pitch information corresponding to each audio frame into an acoustic conversion model to obtain predicted spectrum feature information corresponding to each audio frame; determining an initial loss value corresponding to each audio frame according to the predicted spectrum feature information and the reference spectrum feature information corresponding to each audio frame; determining a weight value corresponding to each initial loss value, and calculating a comprehensive loss value according to the initial loss value and the weight value corresponding to each audio frame; and training and adjusting the acoustic conversion model according to the comprehensive loss value. According to the embodiment of the invention, the accuracy of the acoustic conversion model after training adjustment can be improved to a certain extent.

Description

technical field [0001] The present application relates to the technical field of the Internet, in particular to a method for training an acoustic conversion model, a terminal and a storage medium. Background technique [0002] In recent years, song synthesis technology has been attracting attention from all walks of life. The greatest convenience of this technology is that it can synthesize the audio of songs that someone has not sung. For example, if a user wants to listen to Xu Song's song "Ordinary Road", but Xu Song has not sung the song "Ordinary Road", the user can first find the song audio "Ordinary Road" sung by Pu Shu. " and the lyrics of the song audio "Ordinary Road", so that the terminal generates Xu Song's song audio "Ordinary Road" based on the song audio "Ordinary Road" sung by Pu Shu and the lyrics of the song audio "Ordinary Road". In the above process, the specific steps for the terminal to generate Xu Song's song audio "Ordinary Road" are: input the lyric...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G10H1/00G10H1/02
CPCG10H1/0025G10H1/02G10H2210/111G10H2210/131G10H2250/311Y02T10/40
Inventor 庄晓滨姜涛胡鹏吴斌黄昕周思瑜
Owner TENCENT MUSIC ENTERTAINMENT TECH SHENZHEN CO LTD