Audio identification during performance
Methods and apparatus for audio identification during a performance are disclosed herein. An example apparatus includes at least one memory and at least one processor to transform a segment of audio into a log-frequency spectrogram based on a constant Q transform using a logarithmic frequency resolution, transform the log-frequency spectrogram into a binary image, each pixel of the binary image corresponding to a time frame and frequency channel pair, each frequency channel representing a corresponding quarter tone frequency channel in a range from C3-C8, generate a matrix product of the binary image and a plurality of reference fingerprints, normalize the matrix product to form a similarity matrix, select an alignment of a line in the similarity matrix that intersects one or more bins in the similarity matrix with the largest calculated Hamming similarities, and select a reference fingerprint based on the alignment.
1. An apparatus to identify audio, the apparatus comprising:
at least one memory; and
at least one processor to at least:
transform a segment of audio into a log-frequency spectrogram based on a constant Q transform using a logarithmic frequency resolution;
transform the log-frequency spectrogram into a binary image, each pixel of the binary image corresponding to a time frame and frequency channel pair, each frequency channel representing a corresponding quarter tone frequency channel in a range from musical note C3 to musical note C8;
generate a matrix product of the binary image and a plurality of reference fingerprints;
normalize, based on a number of frequency ranges represented in the binary image, the matrix product to form a similarity matrix;
select an alignment of a line in the similarity matrix that intersects one or more bins in the similarity matrix with the largest calculated Hamming similarities; and
select a reference fingerprint based on the alignment.
2. The apparatus of claim 1 , wherein the processor selects the alignment without using a hash function.
3. The apparatus of claim 1 , wherein the line has an angle within a range of forty to fifty degrees in the similarity matrix.
4. The apparatus of claim 3 , wherein the angle of the line represents a tempo deviation between the segment of audio and the audio represented by the corresponding reference fingerprints.
5. The apparatus of claim 1 , wherein to compute the matrix product, the processor introduces one or more pitch shifts to the binary image.
6. The apparatus of claim 5 , wherein the one or more pitch shifts correspond to key variations in a range of negative five semitones to positive five semitones.
7. The apparatus of claim 1 , wherein the processor selects the alignment using a Hough Transform.
8. The apparatus of claim 1 , wherein to the processor computes the matrix product using a Hamming similarity.
9. The apparatus of claim 1 , wherein the processor binarizes the similarity matrix based on a threshold value.
10. An apparatus to identify audio, the apparatus comprising:
means for transforming a segment of audio into a log-frequency spectrogram based on a constant Q transform using a logarithmic frequency resolution, the means for transforming to transform the log-frequency spectrogram into a binary image, each pixel of the binary image corresponding to a time frame and frequency channel pair, each frequency channel representing a corresponding quarter tone frequency channel in a range from musical note C3 to musical note C8;
means for generating a matrix product of the binary image and a plurality of reference fingerprints;
means for normalizing, based on a number of frequency ranges represented in the binary image, the matrix product to form a similarity matrix; and
means for selecting an alignment of a line in the similarity matrix that intersects one or more bins in the similarity matrix with the largest calculated Hamming similarities, the means for selecting to select a reference fingerprint based on the alignment.
11. The apparatus of claim 10 , wherein the means for selecting comprises selecting the alignment without using a hash function.
12. The apparatus of claim 10 , wherein the line has an angle within a range of forty to fifty degrees in the similarity matrix.
13. The apparatus of claim 12 , wherein the angle of the line represents a tempo deviation between the segment of audio and the audio represented by the corresponding reference fingerprints.
14. The apparatus of claim 10 , wherein the means for generating comprises introducing one or more pitch shifts to the binary image.
15. The apparatus of claim 14 , wherein the one or more pitch shifts correspond to key variations in a range of negative five semitones to positive five semitones.
16. The apparatus of claim 10 , wherein the means for selecting comprises selecting the alignment using a Hough Transform.
17. The apparatus of claim 10 , wherein the means for generating comprises generating the matrix product using a Hamming similarity.
18. The apparatus of claim 10 , further including means for binarizing the similarity matrix based on a threshold value.