Voice conversion model training method and device, electronic equipment and storage medium
A technology for speech conversion and model training, applied in speech analysis, speech recognition, instruments, etc., can solve the problems of low model training accuracy, inaccurate training data labels, and large differences in speech features, and achieve the effect of improving model training accuracy.
Active Publication Date: 2021-11-30
SHENZHEN RAISOUND TECH
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] However, in the current training process of the speech conversion model, since ordinary speech and noise speech are not recorded at the same time, and the speech features of each frame of speech are quite different, the labels of the training data are inaccurate, resulting in low training accuracy of the model
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0052] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments It is a part of the embodiments of this application, but not all of them. Based on the embodiments in the present application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present application.
[0053] figure 1 It is a schematic flowchart of a speech conversion model training method provided in the embodiment of the present application.
[0054] S1. Obtain at least two kinds of voices simultaneously recorded by the user according to the preset content in the first state, the at least two voices include the first voice recorded...
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
Login to View More Abstract
The invention relates to the technical field of artificial intelligence and discloses a voice conversion model training method and device, electronic equipment and a storage medium. The method comprises the following steps of simultaneously acquiring voices of a user in a first state by using a first recording device for non-acoustic pickup and a second recording device for acoustic pickup to respectively obtain a first voice and a second voice; obtaining the voice of the user in the second state by using the first recording device and the second recording device, and respectively obtaining a first noise voice and a second noise voice; aligning the converted frequency spectrums of the first noise voice and the second noise voice; and aligning the second voice with the second noise voice according to a spectrum alignment relationship between the first voice and the first noise voice, and training a voice deep learning model by using the aligned second voice and second noise voice to obtain a voice conversion model. The method is advantaged in that training precision of the voice conversion model can be improved.
Description
technical field [0001] The present application relates to the technical field of artificial intelligence, and in particular to a speech conversion model training method, device, electronic equipment and storage medium. Background technique [0002] With the development of speech recognition technology, a variety of speech data, such as ordinary speech data or noise data with added noise, is needed to train the recognition effect of different speech. However, it is difficult to obtain speech data. In order to better obtain different types of Speech data needs to be converted to speech data, such as denoising noise data and transforming it into ordinary speech, or adding noise to ordinary speech and converting it into noise speech. Currently, the conversion of speech data requires the training of a speech conversion model for conversion. [0003] However, in the training process of the current speech conversion model, ordinary speech and noise speech are not recorded at the sa...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/02G10L15/06G10L15/20
Inventor 黄石磊程刚陈诚廖晨
Owner SHENZHEN RAISOUND TECH