Robust authentication of digital audio

1 . A method of authenticating digital audio, the method comprising:

receiving a digital audio file, the digital audio file decomposed by linear predictive coding (LPC) analysis into a spectral envelope signal and an excitation signal;

determining a psychoacoustic strength factor, based on a multiplication factor for a watermark, ensuring that watermark energy remains below a masking curve, the masking curve associated with a threshold of human hearing;

generating a first watermark using a first key, the first key derived from the spectral envelope signal, wherein the first watermark is band-limited to a first bandwidth, and a watermark energy of the first watermark is below the psychoacoustic strength factor, the first watermark comprising a spread-spectrum watermark configured for embedding in a higher-frequency non-overlapping bandwidth using a subband decomposition path;

generating a second watermark using a second key, the second key derived from the excitation signal, wherein the second watermark is band-limited to a second bandwidth, a watermark energy of the second watermark is below the psychoacoustic strength factor, and wherein the second bandwidth does not overlap with the first bandwidth, the second watermark comprising a self-correlated watermark configured for embedding in a lower-frequency non-overlapping bandwidth using a discrete cosine transform (DCT) path;

embedding the first watermark into a segment of the digital audio file using a spread-spectrum embedding process configured to operate on subband-decomposed audio; and

embedding the second watermark into the segment of the digital audio file using a self-correlated embedding process configured to operate on DCT-domain excitation components.

2 . The method of claim 1 , wherein the first bandwidth has a lower frequency limit above 5 kilohertz (KHz) and the second bandwidth has an upper frequency limit below 5 KHz.

3 . The method of claim 1 , further comprising:

determining a first watermark score of the first watermark using the first key;

determining a second watermark score of the second watermark using the second key; and

based on at least the first watermark score and the second watermark score, determining a probability that the digital audio file is watermarked.

4 . The method of claim 3 , further comprising:

determining, using a machine learning (ML) component, a third watermark score of the segment of the digital audio file, wherein determining the probability that the digital audio file is watermarked is based on the first watermark score, the second watermark score, and the third watermark score.

5 . A system for authenticating digital audio, the system comprising:

a processor; and a computer-readable medium storing instructions that are operative upon execution by the processor to:

receive a digital audio file, the digital audio file decomposed by linear predictive coding (LPC) analysis into a spectral envelope signal and an excitation signal;

determine a psychoacoustic strength factor, based on a multiplication factor for a watermark, ensuring that watermark energy remains below a masking curve, the masking curve associated with a threshold of human hearing;

generate a first watermark using a first key, the first key derived from the spectral envelope signal, wherein the first watermark is band-limited to a first bandwidth, and a watermark energy of the first watermark is below the psychoacoustic strength factor, the first watermark comprising a spread-spectrum watermark configured for embedding in a higher-frequency non-overlapping bandwidth using a subband decomposition path;

generate a second watermark using a second key, the second key derived from the excitation signal, wherein the second watermark is band-limited to a second bandwidth, a watermark energy of the second watermark is below the psychoacoustic strength factor, and wherein the second bandwidth does not overlap with the first bandwidth, the second watermark comprising a self-correlated watermark configured for embedding in a lower-frequency non-overlapping bandwidth using a discrete cosine transform (DCT) path;

embed the first watermark into a segment of the digital audio file using a spread-spectrum embedding process configured to operate on subband-decomposed audio; and

embed the second watermark in the segment of the digital audio file using a self-correlated embedding process configured to operate on DCT-domain excitation components.

6 . The system of claim 5 , wherein the first bandwidth has a lower frequency limit above 5 kilohertz (KHz) and the second bandwidth has an upper frequency limit below 5 KHz.

7 . The system of claim 5 , wherein the instructions are further operative to:

determine a first watermark score of the first watermark using the first key;

determine a second watermark score of the second watermark using the second key; and

based on at least the first watermark score and the second watermark score, determine a probability that the digital audio file is watermarked.

8 . The system of claim 7 , wherein the instructions are further operative to:

determine, using a machine learning (ML) component, a third watermark score of the segment of the digital audio file, wherein determining the probability that the digital audio file is watermarked comprises, based on at least the first watermark score, the second watermark score, and the third watermark score, determining the probability that the digital audio file is watermarked.

9 . One or more computer storage devices having computer-executable instructions stored thereon, which, on execution by a computer, cause the computer to perform operations comprising:

receiving a digital audio file, the digital audio file decomposed by linear predictive coding (LPC) analysis into a spectral envelope signal and an excitation signal;

determining a first watermark score of a segment of the digital audio file for a first watermark using a first key, the first key derived from the spectral envelope signal, wherein the first watermark is band-limited to a first bandwidth, and a watermark energy of the first watermark is below a psychoacoustic strength factor, the psychoacoustic strength factor based on a multiplication factor for a watermark, ensuring that watermark energy remains below a masking curve, the masking curve associated with a threshold of human hearing, the first watermark comprising a spread-spectrum watermark configured for embedding in a higher-frequency non-overlapping bandwidth using a subband decomposition path;

determining a second watermark score of the segment of the digital audio file for a second watermark using a second key, the second key derived from the excitation signal, and a watermark energy of the second watermark is below the psychoacoustic strength factor, wherein the second watermark is band-limited to a second bandwidth, and wherein the second bandwidth does not overlap with the first bandwidth, using LPC synthesis, the second watermark comprising a self-correlated watermark configured for embedding in a lower-frequency non-overlapping bandwidth using a discrete cosine transform (DCT) path;

based on at least the first watermark score and the second watermark score, determining a probability that the digital audio file is watermarked; and

based on at least determining the probability that the digital audio file is watermarked, generating a report indicating whether the digital audio file is watermarked.

10 . The one or more computer storage devices of claim 9 , wherein the first bandwidth has a lower frequency limit above 5 kilohertz (KHz) and the second bandwidth has an upper frequency limit below 5 KHz.

11 . The one or more computer storage devices of claim 9 , wherein the operations further comprise:

determining a third watermark score of a third watermark using a third key, wherein the third watermark is band-limited to a third bandwidth, wherein the third bandwidth does overlap with the first bandwidth or the second bandwidth and wherein determining the probability that the digital audio file is watermarked comprises, based on at least the first watermark score, the second watermark score, and the third watermark score, determining the probability that the digital audio file is watermarked.

12 . The one or more computer storage devices of claim 9 , wherein the operations further comprise:

generating the first watermark using the first key;

generating the second watermark using the second key;

embedding the first watermark into the segment of the digital audio file; and

embedding the second watermark into the segment of the digital audio file.