AI-generated music derivative works
Program instructions on a non-transitory computer-readable medium are executable by at least one processor to receive predetermined content, a request to transform the predetermined content into a derivative work, and one or more user-specified parameters. Generative artificial intelligence is used to generate the derivative work as a function of the predetermined content and the one or more user-specified parameters. The derivative work includes audio, video, or images. Based on a content approval machine learning model, a determination is made as to whether the derivative work is approved based on a content approval score. The content approval score is based on at least one content owner preference and the derivative work. In response to determining the content approval score exceeds a threshold, a digital watermark is applied to the derivative work, an authorization server is configured to govern use of the derivative work, and access to the derivative work is provided.
1 . A non-transitory computer-readable medium having stored thereon program instructions that, when executed by at least one processor, cause the at least one processor to:
receive predetermined content from a database server;
receive a request to transform the predetermined content into a derivative work;
receive one or more user-specified parameters for the derivative work;
generate the derivative work as a function of the predetermined content and the one or more user-specified parameters, using generative artificial intelligence, wherein generating the derivative work comprises transforming the predetermined content into the derivative work comprising audio, video or images, based on an embedding for the one or more user-specified parameters;
determine if the derivative work is approved based on a content approval machine learning model comprising a neural network configured to determine a content approval score as a function of at least one content owner preference and the derivative work, and
in response to determining the content approval score is greater than a threshold:
apply a digital watermark to the approved derivative work;
configure an authorization server to govern use of the approved derivative work based on the digital watermark; and
provide access to the approved derivative work.
2 . The non-transitory computer-readable medium of claim 1 , wherein the content comprises music.
3 . The non-transitory computer-readable medium of claim 1 , wherein the content comprises audio.
4 . The non-transitory computer-readable medium of claim 3 , wherein the audio comprises a human voice sound.
5 . The non-transitory computer-readable medium of claim 4 , wherein the program instructions further cause the at least one processor to detect the human voice based on a technique comprising autocorrelation.
6 . The non-transitory computer-readable medium of claim 5 , wherein the autocorrelation further comprises frequency domain autocorrelation.
7 . The non-transitory computer-readable medium of claim 3 , wherein the audio comprises a musical instrument sound.
8 . The non-transitory computer-readable medium of claim 1 , wherein the one or more user-specified parameters is determined based on an interview with a user.
9 . The non-transitory computer-readable medium of claim 8 , wherein the interview with the user is performed by a chatbot.
10 . The non-transitory computer-readable medium of claim 8 , wherein at least one user-specified parameter of the one or more user-specified parameters is determined based on matching a response from the user with a semantically similar predetermined theme identified by a Large Language Model (LLM) as a function of the response from the user.
11 . The non-transitory computer-readable medium of claim 10 , wherein the predetermined theme is pre-approved by the content owner.
12 . The non-transitory computer-readable medium of claim 1 , wherein the generative artificial intelligence comprises a diffusion model.
13 . The non-transitory computer-readable medium of claim 12 , wherein the diffusion model is a latent diffusion model.
14 . The non-transitory computer-readable medium of claim 12 , wherein the program instructions further cause the at least one processor to encode the content to a latent space using an encoder network.
15 . The non-transitory computer-readable medium of claim 14 , wherein the encoder network comprises a convolutional neural network (CNN) configured to extract mel-frequency cepstral coefficients (MFCCs) from the content.
16 . The non-transitory computer-readable medium of claim 14 , wherein the encoder network comprises a CNN configured to extract a spatial or temporal feature from the content.
17 . The non-transitory computer-readable medium of claim 14 , wherein the encoder network comprises a recurrent neural network (RNN) or a transformer, configured to extract a word embedding from the content.
18 . The non-transitory computer-readable medium of claim 14 , wherein the program instructions further cause the at least one processor to decode the content from the latent space, using a decoder network.
19 . The non-transitory computer-readable medium of claim 1 , wherein the program instructions further cause the at least one processor to determine a text embedding identifying an item, using a CLIP model.
20 . The non-transitory computer-readable medium of claim 1 , wherein the program instructions further cause the at least one processor to convert the one or more user-specified parameters to a text embedding in a shared latent space.
21 . The non-transitory computer-readable medium of claim 1 , wherein applying the digital watermark further comprises embedding the digital watermark in the derivative work.
22 . The non-transitory computer-readable medium of claim 21 , wherein the program instructions further cause the at least one processor to embed the digital watermark in a frequency-domain representation.
23 . The non-transitory computer-readable medium of claim 1 , wherein the program instructions further cause the at least one processor to update the derivative work with a new digital watermark that is valid for a limited time and to provide access to the updated derivative work.
24 . The non-transitory computer-readable medium of claim 1 , wherein governing use of the approved derivative work comprises configuring a tracking system to determine authenticity of the derivative work, verified as a function of the digital watermark by the tracking system.
25 . The non-transitory computer-readable medium of claim 1 , wherein governing use of the approved derivative work comprises validating user requests for access to the derivative work authorized as a function of the digital watermark.
26 . The non-transitory computer-readable medium of claim 1 , wherein governing use of the approved derivative work comprises automatically requesting an automated payment via a smart contract execution triggered based on use of the derivative work detected as a function of the digital watermark.
27 . The non-transitory computer-readable medium of claim 1 , wherein governing use of the approved derivative work comprises revoking access to the derivative work upon determining a time-sensitive watermark has expired.
28 . The non-transitory computer-readable medium of claim 1 , wherein the one or more user-specified parameters comprise a requested theme.
29 . The non-transitory computer-readable medium of claim 1 , wherein the one or more user-specified parameters are received from a content derivation platform.
30 . A non-transitory computer-readable medium having stored thereon program instructions that, when executed by at least one processor, cause the at least one processor to:
receive predetermined content from a database server;
receive a request to transform a segment of the predetermined content to create a derivative work;
receive one or more user-specified parameters for the segment of the derivative work;
convert the user-specified parameters for the segment to a text embedding in a latent space;
generate the derivative work as a function of the segment of the predetermined content and the one or more user-specified parameters, using generative artificial intelligence, wherein generating the derivative work comprises transforming the segment of the predetermined content to create the derivative work comprising one or more of audio, video or images, based on the text embedding for the one or more user-specified parameters;
determine if the derivative work is approved based on a content approval machine learning model comprising a neural network configured to determine a content approval score as a function of at least one content owner preference and the derivative work, and
in response to determining the content approval score is greater than a threshold:
apply a digital watermark to the approved derivative work;
configure an authorization server to govern use of the approved derivative work based on the digital watermark; and
provide access to the approved derivative work.