Audio detection method and apparatus, computer device, and readable storage medium

Patent metadata

Field	Value
Publication number	US-12183315-B2
Application number	US-202217974452-A
Country	US
Kind code	B2
Filing date	Oct 26, 2022
Priority date	Nov 25, 2020
Publication date	Dec 31, 2024
Grant date	Dec 31, 2024

Abstract

This application provide an audio detection method performed by a computer device. The method includes: acquiring a target time point and a reference point of the target time point from target audio data; performing energy evaluation on the target time point according to an audio amplitude value of the target time point to obtain an energy evaluation value of the target time point; performing energy evaluation on the reference point according to an audio amplitude value of the reference point to obtain an energy evaluation value of the reference point; performing accuracy verification on the target time point according to the energy evaluation value of the target time point and the energy evaluation value of the reference point; and if the accuracy verification on the target time point succeeds, adding the target time point as a target stress point into a target stress point set.

First claim

What is claimed is:

1. An audio detection method performed by a computer device, the method comprising:
acquiring a target time point and a reference point of the target time point from target audio data, the target audio data comprising a plurality of time points and an audio amplitude value for each time point, and the reference point referring to a time point with a time difference from the target time point being less than a first difference threshold;
performing energy evaluation on the target time point according to an audio amplitude value of the target time point to obtain an energy evaluation value of the target time point;
performing energy evaluation on the reference point according to an audio amplitude value of the reference point to obtain an energy evaluation value of the reference point; and
performing accuracy verification on the target time point according to the energy evaluation value of the target time point and the energy evaluation value of the reference point, including:
calculating an energy mean of the energy evaluation value of the reference point and the energy evaluation value of the target time point;
determining a maximum energy evaluation value from the energy evaluation value of the target time point and the energy evaluation value of the reference point; and
when a difference between the maximum energy evaluation value and the energy mean is greater than a threshold:
determining that the accuracy verification on the target time point succeeds; and
adding the target time point as a target stress point into a target stress point set.

2. The method according to claim 1 , wherein performing the accuracy verification on the target time point according to the energy evaluation value of the target time point and the energy evaluation value of the reference point further comprises:
when the difference between the maximum energy evaluation value and the energy mean is not greater than the threshold, determining that the accuracy verification on the target time point fails.

3. The method according to claim 1 , wherein:
the plurality of time points are arranged in chronological order; and
performing the energy evaluation on the target time point according to the audio amplitude value of the target time point to obtain the energy evaluation value of the target time point comprises:
acquiring a plurality of associated points of the target time point from the plurality of time points, and calculating an audio energy value of the target time point by using an audio energy function according to audio amplitude values of the associated points and the audio amplitude value of the target time point, the associated point referring to a time point with a time difference from the target time point being less than a second difference threshold;
acquiring a preceding point of the target time point from the plurality of time points, the preceding point comprising c time points selected forward in sequence based on an arrangement position of the target time point in the plurality of time points, c being a positive integer;
calculating an audio energy change value of the target time point by using an audio energy change function according to the audio energy value of the target time point and audio energy values of time points in the preceding point; and
performing weighted summation on the audio energy value and the audio energy change value to obtain the energy evaluation value of the target time point.

4. The method according to claim 3 , wherein the calculating an audio energy value of the target time point by using an audio energy function according to audio amplitude values of the associated points and the audio amplitude value of the target time point comprises:
performing a square operation on the audio amplitude value of the target time point to obtain an initial energy value of the target time point; performing a square operation on the audio amplitude value of each associated point to obtain an initial energy value of each associated point; and
performing a mean operation on the initial energy value of the target time point and initial energy values of the associated points to obtain the audio energy value of the target time point.

5. The method according to claim 4 , wherein the performing a mean operation on the initial energy value of the target time point and initial energy values of the associated points to obtain the audio energy value of the target time point comprises:
performing a mean operation on the initial energy value of the target time point and the initial energy values of the associated points to obtain an intermediate energy value; and
denoising the intermediate energy value to obtain the audio energy value of the target time point.

6. The method according to claim 3 , wherein the calculating an audio energy change value of the target time point by using an audio energy change function according to the audio energy value of the target time point and audio energy values of time points in the preceding point comprises:
calculating a sum of the audio energy values of the time points in the preceding point;
acquiring a reference value, and calculating a difference between the sum of the audio energy values and c times the audio energy value of the target time point;
using a maximum value in the reference value and the calculated difference through calculation as an initial energy change value of the target time point; and
determining the audio energy change value of the target time point according to the initial energy change value of the target time point.

7. The method according to claim 6 , wherein the determining the audio energy change value of the target time point according to the initial energy change value of the target time point comprises:
acquiring initial energy change values of time points in the target audio data;
determining a plurality of peaks from the initial energy change values of the time points, each peak referring to an initial energy change value of a peak time point in the target audio data, and the peak time point satisfying the following condition: the initial energy change value of the peak time point being greater than an initial energy change value of each of two time points respectively on left and right sides of the peak time point and adjacent to the peak time point; and
normalizing the initial energy change value of the target time point by using a mean of the plurality of peaks to obtain the audio energy change value of the target time point.

8. The method according to claim 7 , wherein the normalizing the initial energy change value of the target time point by using a mean of the plurality of peaks to obtain the audio energy change value of the target time point comprises:
acquiring audio energy values of time points, and determining a minimum audio energy value from the audio energy values of the time points; and
performing contraction on the initial energy change value of the target time point by using the mean of the plurality of peaks and the minimum audio energy value to obtain the audio energy change value of the target time point.

9. The method according to claim 3 , further comprising:
before adding the target time point as a target stress point into a target stress point set:
selecting, from absolute values of audio amplitude values of the associated points and an absolute value of the audio amplitude value of the target time point, a maximum absolute value as a local maximum amplitude value of the target time point; and
when the local maximum amplitude value of the target time point is greater than a first amplitude threshold, adding the target time point as a target stress point into the target stress point set.

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Huang Zhengyue
• Shi Xintian

Classifications

• G10H2250/235

  Fourier transform; Discrete Fourier Transform [DFT]; Fast Fourier Transform [FFT] · CPC title

• G10H2210/031

  Musical analysis, i.e. isolation, extraction or identification of musical elements or musical parameters from a raw acoustic signal or from an encoded audio signal · CPC title

• G10H1/40

  Rhythm · CPC title

• G10L25/51

  for comparison or discrimination · CPC title

• G10H2210/051

  for extraction or detection of onsets of musical sounds or notes, i.e. note attack timings · CPC title