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

Training method and system of singing sound synthesis model and singing sound synthesis method

A training method and singing technology, applied in speech synthesis, speech analysis, instruments, etc., can solve the problems of low training and synthesis efficiency, lack of synthesis, etc.

Pending Publication Date: 2020-01-31
PING AN TECH (SHENZHEN) CO LTD
View PDF0 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of this, the purpose of the embodiments of the present invention is to provide a training method, system, computer equipment and computer-readable storage medium of a singing voice synthesis model, which can effectively solve the technical problems of low training and synthesis efficiency and lack of flexibility in synthesis

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
  • Training method and system of singing sound synthesis model and singing sound synthesis method
  • Training method and system of singing sound synthesis model and singing sound synthesis method
  • Training method and system of singing sound synthesis model and singing sound synthesis method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] refer to figure 1 , shows a flow chart of the steps of the training method of the singing voice synthesis model in Embodiment 1 of the present invention. It can be understood that the flowchart in this method embodiment is not used to limit the sequence of execution steps. details as follows.

[0057] Step S100, acquiring multiple singing voice data of multiple songs, and building a training database based on the multiple singing voice data and multiple music scores corresponding to the multiple songs.

[0058] The singing voice data is recorded audio data, and generally speaking, the singing voice data includes the singing voice of a designated person (professional singer) and the voice of an accompanying instrument. But, when there is no accompaniment instrument, then described singing voice data is the singing voice sent by the appointed person.

[0059] Exemplarily, the singing voice of a designated person (professional singer) can be recorded through a recording...

Embodiment 2

[0092] read on figure 2, shows a schematic diagram of the program modules of Embodiment 2 of the training system for the singing voice synthesis model of the present invention. In this embodiment, the training system 20 of the singing synthesis model may include or be divided into one or more program modules, one or more program modules are stored in a storage medium, and are executed by one or more processors, To complete the present invention, and can realize the training method of above-mentioned singing voice synthesis model. The program module referred to in the embodiment of the present invention refers to a series of computer program instruction segments capable of completing specific functions, which is more suitable than the program itself to describe the execution process of the training system 20 of the singing voice synthesis model in the storage medium. The following description will specifically introduce the functions of each program module of the present embo...

Embodiment 3

[0102] refer to image 3 , is a schematic diagram of the hardware architecture of the computer device according to Embodiment 3 of the present invention. In this embodiment, the computer device 2 is a device capable of automatically performing numerical calculation and / or information processing according to preset or stored instructions. The computer device 2 may be a rack server, a blade server, a tower server or a cabinet server (including an independent server, or a server cluster composed of multiple servers) and the like. As shown in the figure, the computer device 2 at least includes, but is not limited to, a memory 21, a processor 22, a network interface 23, and a training system 20 for singing voice synthesis models that can communicate with each other through a system bus. in:

[0103] In this embodiment, the memory 21 includes at least one type of computer-readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, ca...

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 embodiment of the invention provides a training method of a singing sound synthesis model. The method comprises the following steps: acquiring a plurality of singing sound data of a plurality of songs, and constructing a training database based on the plurality of singing sound data and a plurality of music scores corresponding to the plurality of songs; segmenting the singing sound data of each song into a plurality of voice frames, segmenting the music score data of each song into a plurality of music score voice elements, and establishing a mapping relationship between each music scorevoice element in each song and the plurality of corresponding voice frames; extracting an acoustic feature from the voice frame corresponding to each music score voice element of each song; and training the singing sound synthesis model according to the music score voice elements of each song and the acoustic feature corresponding to each music score voice element to obtain a trained singing soundsynthesis model. According to the embodiment, the singing sound synthesis model corresponding to a certain singer or certain singers can be obtained by efficient and flexible training through a smallnumber of corpuses.

Description

technical field [0001] Embodiments of the present invention relate to the field of computer data processing, and in particular, relate to a training method, system, computer equipment, computer-readable storage medium, and a singing voice synthesis method for a singing voice synthesis model. Background technique [0002] With the development of Internet and digital storage technology, audio files are mostly recorded and transmitted in digital formats, such as WAV, MP3, MIDI and so on. Audio files in digital format have incomparable advantages in production, storage, and distribution. Creators can compose music through computer equipment and output the production effect of music works, and any modifications to the score can be fed back to the creators in a timely manner, effectively reducing the cycle and labor costs of music production. In recent years, the singing voice synthesis technology has been greatly developed, and the current singing voice synthesis system mainly i...

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): G10L13/02G10L13/04G10L13/08G10L13/10G10L25/03G10L25/18G10L25/24G10L25/30
CPCG10L13/02G10L13/08G10L13/10G10L25/03G10L25/30G10L25/24G10L25/18G10L13/00
Inventor 王健宗
Owner PING AN TECH (SHENZHEN) CO LTD