Methods and apparatus to reduce noise from harmonic noise sources
View Patent ↗Methods, apparatus, systems and articles of manufacture are disclosed to reduce noise from harmonic noise sources. Example methods disclosed herein include determining a first point representing a comparatively large amplitude for a frequency value in an audio sample. Disclosed example methods also include generating a first contour trace of points having amplitude, frequency and phase values within thresholds from the first point of comparatively large amplitude. Example methods include generating a second contour trace of points having amplitude, frequency and phase values within thresholds of a second point of comparatively large amplitude. Disclosed example methods include calculating a parameter for each of the contour traces and determining if the first or second contour traces represent outliers based on the parameters. Disclosed example methods also include, in response to determining the contour to be an outlier contour trace, removing the outlier contour trace from the audio sample.
1. An apparatus to reduce harmonic noise, the apparatus comprising:
a contour tracer to:
determine a first point representing a comparatively large amplitude for a frequency value in an audio sample;
generate a first contour trace of other points having amplitude, frequency and phase values within specified thresholds from the first point, the points of the first contour trace occurring in succession within a specified number of frames from each other or from the first point; and
generate a second contour trace of points having amplitude, frequency, and phase values within thresholds of a second point representing a comparatively large amplitude for a frequency value in the audio sample, the points of the second contour trace occurring in succession within a specified number of frames from each other or from the second point;
a parameter calculator to calculate a first parameter for the first contour trace and a second parameter for the second contour trace;
a classifier to determine if the first and second contour traces represent outliers based on the first and second parameters; and
a subtractor to, in response to determining the first or second contour to be an outlier contour trace, remove the outlier contour trace from the audio sample.
2. The apparatus of claim 1 , wherein the contour tracer is to generate generates a third contour trace of points having amplitude, frequency and phase values within thresholds of a third point representing a comparatively large amplitude for a frequency value in the audio sample, the third point being within a frequency range threshold of a harmonic frequency of the frequency represented by the first point, the points of the third contour trace occurring in succession within a specified number of frames from each other or from the third point.
3. The apparatus of claim 1 , wherein the contour tracer is to determine a point of comparatively large amplitude for a representative number of frequencies in the audio sample and to generate contours for a specified percentage of the points of comparatively large amplitude in the audio sample.
4. The apparatus of claim 1 , wherein the classifier is to determine if the first and second contour traces represent outliers based on a statistical distance from an average of the first and second parameters.
5. The apparatus of claim 1 , wherein removal of outlier contour traces improves a noise-reduced audio signal that is used to generate an accurate audio signature.
6. The apparatus of claim 1 , further including a domain converter to perform a short-time Fourier transform with a specified windowing length and window time frame on the audio sample.
7. The apparatus of claim 6 , wherein the contour tracer is to conclude generating a contour when a specified number of short-time Fourier transform frames have been analyzed without finding a point within the specified thresholds to be part of the first or second contour.
8. A method comprising:
determining, by executing an instruction with a processor, a first point representing a comparatively large amplitude for a frequency value in an audio sample;
generating, by executing an instruction with the processor, a first contour trace of other points having amplitude, frequency and phase values within specified thresholds from the first point, the points of the first contour trace occurring in succession within a specified number of frames from each other or from the first point;
generating, by executing an instruction with the processor, a second contour trace of points having amplitude, frequency and phase values within thresholds of a second point representing a comparatively large amplitude for a frequency value in the audio sample, the points of the second contour trace occurring in succession within a specified number of frames from each other or from the second point;
calculating, by executing an instruction with the processor, a first parameter for the first contour trace and a second parameter for the second contour trace;
determining, by executing an instruction with the processor, if the first or second contour traces represent outliers based on the first and second parameters; and
in response to determining the first or the second contour to be an outlier contour trace, removing, by executing an instruction with the processor, the outlier contour trace from the audio sample.
9. The method of claim 8 , further including generating, by executing an instruction with the processor, a third contour trace of points having amplitude, frequency and phase values within thresholds of a third point representing a comparatively large amplitude for a frequency value in the audio sample, the third point being within a frequency range threshold of a harmonic frequency of the frequency represented by the first point, the points of the third contour trace occurring in succession within a specified number of frames from each other or from the third point.
10. The method of claim 8 , further including determining a point of comparatively large amplitude for a representative number of frequencies in the audio sample and generating contours for a specified percentage of the points of comparatively large amplitude in the audio sample.
11. The method of claim 8 , wherein determining if the first and second contour traces represent outliers is based on a statistical distance from an average of the first and second parameters.
12. The method of claim 8 , wherein removal of outlier contour traces improves a noise-reduced audio signal that is used to generate an accurate audio signature.
13. The method of claim 8 , wherein a short-time Fourier transform with a specified windowing length and window time frame is performed on the audio sample.
14. The method of claim 13 , wherein the generating of the first and second contour concludes when a specified number of short-time Fourier transform frames have been analyzed without finding a point within the specified thresholds to be part of the first or second contour.
15. A non-transitory computer readable storage medium comprising computer readable instructions that, when executed, cause a processor to at least:
determine, by executing an instruction with a processor, a first point representing a comparatively large amplitude for a frequency value in an audio sample;
generate, by executing an instruction with the processor, a first contour trace of other points having amplitude, frequency and phase values within specified thresholds from the first point, the points of the first contour trace occurring in succession within a specified number of frames from each other or from the first point;
generate, by executing an instruction with the processor, a second contour trace of points having amplitude, frequency and phase values within thresholds of a second point representing a comparatively large amplitude for a frequency value in the audio sample, the points of the second contour trace occurring in succession within a specified number of frames from each other or from the second point;
calculate, by executing an instruction with the processor, a first parameter for the first contour trace and a second parameter for the second contour trace;
determine, by executing an instruction with the processor, if the first or second contour traces represent outliers based on the first and second parameters; and
in response to determining the first or the second contour to be an outlier contour trace, remove, by executing an instruction with the processor, the outlier contour trace from the audio sample.
16. The non-transitory computer readable storage medium of claim 15 , further including generating, by executing an instruction with the processor, a third contour trace of points having amplitude, frequency and phase values within thresholds of a third point representing a comparatively large amplitude for a frequency value in the audio sample, the third point being within a frequency range threshold of a harmonic frequency of the frequency represented by the first point, the points of the third contour trace occurring in succession within a specified number of frames from each other or from the third point.
17. The non-transitory computer readable storage medium of claim 15 , wherein the instructions, when executed, cause the processor to determine a point of comparatively large amplitude for a representative number of frequencies in the audio sample and to generate contours for a specified percentage of the points of comparatively large amplitude in the audio sample.
18. The non-transitory computer readable storage medium of claim 15 , wherein determining if the first and second contour traces represent outliers is based on a statistical distance from an average of the first and second parameters.
19. The non-transitory computer readable storage medium of claim 15 , wherein removal of outlier contour traces improves a noise-reduced audio signal that is used to generate an accurate audio signature.
20. The non-transitory computer readable storage medium of claim 15 , wherein a short-time Fourier transform with a specified windowing length and window time frame is performed on the audio sample.