Training method of high-frequency signal reconstruction model and high-frequency signal reconstruction method and device
A high-frequency signal and training method technology, applied in the field of signal processing, can solve the problems of vocal signal quality limitation, affecting the integrity of the sound signal, and the lack of high-frequency signals of the sound signal.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0074] see figure 1 , is a schematic flowchart of a method for training a high-frequency signal reconstruction model provided by an embodiment of the present disclosure. Such as figure 1 As shown, the method mainly includes the following steps:
[0075] S101. Perform low-pass filtering of the first frequency and the second frequency on the audio signal used for model training, respectively, to obtain a sample signal corresponding to the first frequency and a target signal corresponding to the second frequency, wherein the first frequency A frequency is less than the second frequency.
[0076] The training method provided in this embodiment is used for training a high-frequency signal reconstruction model capable of reconstructing a high-frequency part of a signal. The provided high-frequency signal reconstruction model can be suitable for reconstructing the high-frequency part of the human voice signal, especially the human voice signal collected and / or transmitted by the b...
Embodiment 2
[0161] Figure 5 It is a schematic flowchart of a high-frequency signal reconstruction method provided by an embodiment of the present disclosure. The high-frequency signal reconstruction method provided in this embodiment is applied to computer equipment, and the computer equipment is loaded with the high-frequency signal reconstruction model trained by the high-frequency signal reconstruction model training method described in the above-mentioned embodiments. Such as Figure 5 As shown, the method mainly includes the following steps:
[0162] S501. Receive an audio signal to be processed;
[0163] S502. Input the audio signal into a high-frequency signal reconstruction model, and use the high-frequency signal reconstruction model to reconstruct a high-frequency time-domain signal of the audio signal.
[0164] This embodiment further limits the process of reconstructing high-frequency signals using the high-frequency signal reconstruction model trained in the above embodim...
Embodiment 3
[0168] with the above figure 1 The method embodiment shown corresponds to, Image 6 A block diagram of a training device for a high-frequency signal reconstruction model provided by an embodiment of the present disclosure. Such as Image 6 As shown, the training device 600 of the high-frequency signal reconstruction model includes:
[0169] The sampling module 601 is configured to perform low-pass filtering of the first frequency and the second frequency on the audio signal used for model training, respectively, to obtain a sample signal corresponding to the first frequency and a target signal corresponding to the second frequency, wherein , the first frequency is less than the second frequency;
[0170] A transformation module 602, configured to obtain sample characteristic parameters according to the sample signal, and obtain target characteristic parameters according to the target signal;
[0171] The training module 603 is configured to input the sample characteristic ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More