Music and digital rights management system and method

Abstract

A method and associated system for matching digital work metadata and delivering the digital work metadata to one or more digital service providers includes modifying one or more digital work metadata files, modifying the one or more digital work metadata files includes removing non-critical data or segmented wrong data, or performing language translation; the one or more digital work metadata files are reformatted so as to be compatible with AI matching operation based on a transformer model; performing a chunk grouping operation on the one or more digital work metadata files, in which data associated with the one or more digital work metadata files is grouped into chunks and analyzed to obtain one or more pairs of data records; executing the AI matching operation based on the transformer model to determine whether each pair in the one or more pairs of data records comprises a matched pair of the data records; and transmitting the output data from the artificial intelligence matching operation based on the transformer model to the one or more digital service providers.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to U.S. non-provisional application serial number 18 / 117,575, filed March 6, 2023, which is a partial continuation of U.S. non-provisional patent application number 16 / 799,201, filed February 24, 2020, which claims priority to U.S. provisional patent application number 62 / 809,590, filed February 23, 2019, the disclosure and teachings of which are incorporated herein by reference. Technical Field

[0003] This invention relates to systems and methods for music and digital rights management. More specifically, this invention relates to systems and methods for transforming, matching, and delivering metadata of music and digital works to associated recording usage data of a digital service provider (DSP). Background Technology

[0004] Given the pervasive nature of the internet today, actively identifying and collecting all royalties owed to music rights holders is a daunting task. Therefore, a system and associated methods are needed to help music rights holders and other digital rights holders find and collect their owed royalties. Summary of the Invention

[0005] Generally speaking, in one aspect, the present invention is characterized by a method for matching digital work metadata and delivering the digital work metadata to one or more digital service providers, the method comprising: modifying one or more digital work metadata files under the control of one or more processors configured with executable instructions, wherein the modification of the one or more digital work metadata files includes removing non-critical data or data with segmentation errors, or performing language translation; reformatting the one or more digital work metadata files to be compatible with an AI matching operation based on a transformer model; performing a block grouping operation on the one or more digital work metadata files, wherein data associated with the one or more digital work metadata files is grouped into blocks and analyzed to obtain one or more pairs of data records; performing the AI ​​matching operation based on a transformer model on the one or more pairs of data records to determine whether each of the one or more pairs of data records includes a matching pair of data records; and transmitting the output data from the AI ​​matching operation based on a transformer model to the one or more digital service providers.

[0006] Embodiments of the present invention may include one or more of the following features. The one or more digital work metadata files may be audio recording metadata files. The transformer-based AI matching operation may include matching one or more basic criteria, including work title, author, recording title, or recording artist. The output data may be transmitted to the one or more digital service providers via one or more third-party application programming interfaces.

[0007] The method may further include transferring the output data to a rights management dashboard before transmitting it to the one or more digital service providers. The method may also include converting the output data into a data standard format compatible with each of the one or more digital service providers before transmitting it to them. The method may further include performing a second AI matching operation on the pair or more pairs of data records, based on the Dedupe Python library. This transformer-based AI matching operation can be performed on cloud infrastructure.

[0008] Generally speaking, in another aspect, the present invention is characterized by a system configured to match digital work metadata and deliver the digital work metadata to one or more digital service providers, the system comprising: one or more processors; one or more non-transitory computer-readable media; and one or more modules maintained on the one or more non-transitory computer-readable media, the one or more modules causing the one or more processors to perform operations when executed by the one or more processors, the operations including: modifying one or more digital work metadata files, wherein the modification of the one or more digital work metadata files includes removing non-critical data or data with segmentation errors, or performing language translation; reformatting the one or more digital work metadata files to be compatible with an AI matching operation based on a transformer model; performing a block grouping operation on the one or more digital work metadata files, wherein data associated with the one or more digital work metadata files is grouped into blocks and analyzed to obtain one or more pairs of data records; performing the AI ​​matching operation based on a transformer model on the one or more pairs of data records to determine whether each of the one or more pairs of data records includes a matching pair of data records; and transmitting the output data from the AI ​​matching operation based on a transformer model to the one or more digital service providers.

[0009] Embodiments of the present invention may include one or more of the following features. The one or more digital work metadata files may be audio recording metadata files. The transformer-based AI matching operation may include matching one or more basic criteria, including work title, author, recording title, or recording artist. The output data may be transmitted to the one or more digital service providers via one or more third-party application programming interfaces.

[0010] The system may also include the following additional operations: transferring the output data to the rights management dashboard before transmission to the one or more digital service providers. The system may also include the following additional operations: converting the output data into a data standard format compatible with each of the one or more digital service providers before transmission to them. The system may also include the following additional operations: performing a second AI matching operation on the pair or more data records, based on the Dedupe Python library. This transformer-based AI matching operation can be executed on cloud infrastructure. Attached Figure Description

[0011] Figure 1 An operational flowchart illustrating an embodiment of the system and associated method of the present invention is shown;

[0012] Figure 2 An operational flowchart illustrating another embodiment of the present invention utilizing a job scheduling and processing system and associated methods is shown.

[0013] Figure 3 An operational flowchart for the job scheduling and processing system used in this invention is illustrated;

[0014] Figure 4 A first operational flowchart illustrating a system and associated method according to another embodiment of the present invention is shown; and

[0015] Figure 5 Examples Figure 4 The second operational flowchart of the system and associated methods of the embodiment is shown. Detailed Implementation

[0016] This invention relates to a system and associated method for transforming, matching, and delivering associated recording usage data of music and digital works metadata to a DSP. The matching mechanism may include a client data transformation aspect, followed by transmission to an artificial intelligence (AI) entity resolution library. This library can be configured for Python and may be based on open-source libraries such as Dedupe. The system may include a pre-matching analysis mechanism configured to extract positive matches from one or more datasets, such as 1:1 correlations with International Standard Work Code (ISWC), International Standard Recording Code (ISRC), and other authoritative data points, and to transmit non-matching data (if any) to an adapted Dedupe library. After AI processing of the data, the system can extract the results into a specific data standard or file format used to deliver ownership information to an available DSP according to DSP submission requirements. The system can be implemented as an automated software pipeline.

[0017] The system of the present invention may include several aspects. A first aspect is pre-analysis transformation. The system can process metadata files, such as those provided by a customer, to remove any redundant data and / or segment any erroneous or potentially erroneous data. The system can then reformat the file for, for example, merging capabilities with DSP usage reports into a database table.

[0018] The second aspect is pre-analysis matching. The system can utilize metadata files to perform matching analysis with DSP usage report tables. Matching can be performed based on one or more data points, including ISWCs, ISRCs, and other authoritative data points. The resulting dataset can be parsed to determine additional authoritative data points. In a non-limiting example, new ISWCs can be discovered if only ISRCs are available in the metadata file, and new ISRCs can be discovered if only ISWCs are available in the metadata file. Newly discovered authoritative data points can be further matched, adding the results to a table such as recordings of works and links. In one example of such a table, the table could represent data clusters updated through AI matching. Works and recordings that do not form part of a data cluster can undergo AI processing for predictive matching.

[0019] The third aspect is the aforementioned AI matching, specifically via an adapted Dedupe library. This adapted Dedupe library is capable of performing clustering and entity parsing on large datasets, such as those exceeding 25 million rows. The AI ​​matching mechanism can be performed across the entire dataset, including with a training predicate set supporting matching on one or more fundamental criteria, such as work titles & authors and recording titles & artists. This Dedupe library can be configured with one or more statistical analysis-based matching techniques and the training predicate set to generate confidence scores for clustered matching. The generated results can include tables with cluster identifiers, confidence scores, and one or more of the additional data points appended to the original database tables. Clusters can be merged with those created during pre-analysis matching to obtain an affinity linked to the recordings of individual works. Such affinity can be updated when, for example, new work data and DSP usage reports are processed by the system of this invention.

[0020] The fourth aspect is post-analysis transformation, which is configured to transmit the matching file to the rights management dashboard. After approval by the rights administrator, this data can be transformed into a specific data standard, which can be a flat file, such as a comma-separated value (CSV) file. This transformed data can then be ingested by an ingestion pipeline associated with the system to transform the relevant metadata into a DSP-specific format.

[0021] Figure 1 A flowchart illustrating the operation of the system and associated methods according to embodiments of the present invention is provided. A data file (“Customer Data File”) can be uploaded to the rights management dashboard. The data file can be placed in a queue for pre-analysis transformation and for requesting AMAZON WEB SERVICES (AWS) to initiate the AI ​​matching mechanism, as this mechanism can be initiated only on demand. The data file can be transformed and ingested into a PostgreSQL database, and then merged into a DSP usage report table. If the AI ​​matching mechanism is initiated, data is sent to begin AI matching. Specifically, a Dedupe-based pipeline can be performed on the transformed / ingested dataset. The data file can then be transformed from the raw AI output into a readable format for predictive matching and delivered to the rights management dashboard. After approval by the rights administrator, the data can be ingested by the DSP ingestion pipeline. This data is converted into a specific data standard format that supports transmission to all relevant DSPs and is subsequently transmitted to the DSPs, such as via a third-party application programming interface (API).

[0022] Figure 2 An operational flowchart illustrating another embodiment of the system and associated method of the present invention is shown, which is similar to... Figure 1 The embodiments utilize job scheduling and processing systems, such as, but not limited to, Figure 3 The specific implementation of the job scheduling and processing system. Figure 2 Examples of the operations performed within the cloud infrastructure associated with the job scheduling and processing system are illustrated, including pre-analysis transformation, pre-analysis matching, AI matching, and post-analysis transformation, as well as DSP ingestion pipeline operations. Beyond the cloud infrastructure is the rights management dashboard, which can be configured as a graphical user interface (GUI).

[0023] Figure 3 An operational flowchart for a job scheduling and processing system utilized in this invention is illustrated. This job scheduling and processing system is configured to group, monitor, and report on one or more jobs or operation pipelines via cloud infrastructure. Any type of compute job can be scheduled, picked up, and assigned to the cloud by this job scheduling and processing system. The job scheduling and processing system is configured to remain active or available and to scale up the cloud infrastructure as needed or as necessary for requested compute jobs.

[0024] Figure 3 The document describes multiple job or operational pipelines, including those associated with data ingestion, data delivery, and content discovery, including those incorporating AI-based systems. A pipeline may consist of several computational tasks, as indicated by the subscript letters, yielding the desired output upon completion of all relevant tasks. The job scheduling and processing system maintains an understanding of the necessary computational tasks within a particular pipeline via a database table that defines the tasks required to complete the pipeline and the associated steps. These illustrated and described pipelines are exemplary and not limiting.

[0025] like Figure 3 Further examples illustrate aspects that can be controlled or operated by the job scheduling and processing system, including: taking data from the rights management dashboard into a database (e.g., triggered by an end user), transforming data to feed into an AI-based system, transforming the resulting data and taking the resulting data from the AI-based system into a database, transforming data (including the rights management interpretation of the results) into a data format acceptable to all DSPs, and delivering data to the DSP.

[0026] Figures 4 to 5 An operational flowchart illustrating another embodiment of the system and associated method of the present invention is shown. This embodiment is related to... Figure 1The implementation differs in several aspects. First, before initiating the AI ​​matching mechanism, the pipeline data undergoes a chunking mechanism based on tokenization of specific metadata fields, including but not limited to title, author / author, recording title, and recording artist. The chunked data then undergoes a candidate pair generation algorithm that analyzes the chunks to generate groupings of candidate pairs of similar data based on defined thresholds. Candidate pairs can be stored in an easily accessible storage system to enable rapid searching via the system's AI mechanism, which determines whether a candidate pair actually matches (yes / no) without human interpretation or involvement. This AI mechanism can be executed in coordination with the aforementioned Dedupe library, and any one or both sets of results can be utilized as needed, including depending on the specific task.

[0027] Figures 4 to 5 Another aspect of the embodiment is the use of a transformer-based AI in the AI ​​matching pipeline. This transformer-based AI is specifically trained on digital asset metadata, such as music metadata like title, author / author, recording title, artist, etc. The use of the transformer model effectively removes the need for human interpretation of the results and human involvement in decision-making. For example, the model can output a decision about whether two entities are the same with 95% confidence.

[0028] Figures 4 to 5 Another aspect of the embodiment is the inclusion of initial preprocessing in the ingestion pipeline. Prior to the AI ​​matching aspect of this system, various automated preprocessing pipelines can be utilized for the ingested data, including but not limited to language translation, useless data removal or null removal, and record sorting or deduplication.

[0029] At the end of the AI ​​matching pipeline, the matched entities can be stored for future use. Specifically, the matched entities can be stored or represented in a database for use based on associated source data (e.g., Figure 5 The data in set A (as illustrated) will be standardized within the corresponding works or digital assets linked to the matched entities. The storage of matched entities enables the enhancement of relevant data in future analyses, including additional metadata, such as... Figure 5 The data in the illustrated set A, as more data passes through such a system of the present invention, gains improved chunking and matching capabilities over time.

[0030] Figure 4 Provides a structure corresponding to Figure 1 The operation flowchart of the structure. Figure 5 Provided Figure 4An alternative operational flowchart highlights four pipeline aspects: data combination, followed by preprocessing of the combined dataset, then chunking for candidate pair generation, and finally evaluation of the candidate pairs, such as the use of AI based on the Dedupe library or trained Natural Language Processing (NLP) transformer models, as described above.

[0031] Figures 4 to 5 The system, specifically embodied in this, can be used not only to discover relevant matches and deliver them to various digital service providers, but also for royalty reconciliation, i.e., reconciliation discrepancies in surplus data. Relevant surplus data is utilized and extracted, such as... Figure 5 The system of the present invention, as illustrated in set B, which includes a modified preprocessing pipeline, facilitates the matching of internal surplus data to link surpluses to copyright holders.

[0032] The above embodiments and examples are illustrative, and many variations may be introduced thereto without departing from the spirit and scope of this disclosure or the scope of the invention. For example, elements and / or features of different illustrative and exemplary embodiments herein may be combined and / or substituted for each other within the scope of this disclosure. For a better understanding of the invention, its operational advantages, and the specific purposes attained through its use, reference should be made to the accompanying drawings and descriptive details, in which preferred embodiments of the invention are illustrated.

Claims

1. A method for matching digital work metadata and delivering the digital work metadata to one or more digital service providers, the method comprising: Under the control of one or more processors configured with executable instructions, Modify one or more digital work metadata files, wherein modifying one or more digital work metadata files includes removing non-critical data or data with segmentation errors, or performing language translation; Reformat the one or more digital work metadata files to be compatible with AI matching operations based on the transformer model; A block grouping operation is performed on the one or more digital work metadata files, wherein data associated with the one or more digital work metadata files is grouped into blocks and analyzed to obtain one or more pairs of data records; Perform the transformer-based AI matching operation on the one or more pairs of data records to determine whether each of the one or more pairs of data records includes a matching pair of data records; as well as The output data from the AI ​​matching operation based on the transformer model is transmitted to the one or more digital service providers.

2. The method according to claim 1, wherein the one or more digital work metadata files are audio recording metadata files.

3. The method according to claim 1, wherein the AI ​​matching operation based on the transformer model includes matching one or more basic criteria, the one or more basic criteria including the title of the work, the author of the work, the title of the recording, or the recording artist.

4. The method of claim 1, wherein the output data is transmitted to the one or more digital service providers via one or more third-party application programming interfaces.

5. The method of claim 1, further comprising transmitting the output data to a rights management dashboard before transmitting it to the one or more digital service providers.

6. The method of claim 1, further comprising converting the output data into a data standard format compatible with each of the one or more digital service providers before transmitting it to the digital service providers.

7. The method according to claim 1, further comprising performing a second artificial intelligence matching operation on the pair or more pairs of data records, the second artificial intelligence matching operation being based on the Dedupe Python library.

8. The method of claim 1, wherein the AI ​​matching operation based on the transformer model is performed on cloud infrastructure.

9. A system configured to match digital work metadata and deliver the digital work metadata to one or more digital service providers, the system comprising: One or more processors; One or more non-transitory computer-readable media; as well as One or more modules maintained on one or more non-transitory computer-readable media, the one or more modules causing the one or more processors to perform operations when executed by the one or more processors, the operations including: Modify one or more digital work metadata files, wherein modifying one or more digital work metadata files includes removing non-critical data or data with segmentation errors, or performing language translation; Reformat the one or more digital work metadata files to be compatible with AI matching operations based on the transformer model; A block grouping operation is performed on the one or more digital work metadata files, wherein data associated with the one or more digital work metadata files is grouped into blocks and analyzed to obtain one or more pairs of data records; Perform the transformer-based AI matching operation on the one or more pairs of data records to determine whether each pair of the one or more pairs of data records includes a matching pair of data records; and The output data from the AI ​​matching operation based on the transformer model is transmitted to the one or more digital service providers.

10. The system of claim 9, wherein the one or more digital work metadata files are audio recording metadata files.

11. The system of claim 9, wherein the AI ​​matching operation based on the transformer model includes matching one or more basic criteria, the one or more basic criteria including title of work, author of work, title of recording, or recording artist.

12. The system of claim 9, wherein the output data is transmitted to the one or more digital service providers via one or more third-party application programming interfaces.

13. The system of claim 9, further comprising the additional operation of transmitting the output data to a rights management dashboard before transmitting it to the one or more digital service providers.

14. The system of claim 9, further comprising the additional operation of converting the output data into a data standard format compatible with each of the one or more digital service providers before transmission to them.

15. The system of claim 9, further comprising the following additional operation: performing a second artificial intelligence matching operation on the pair or more pairs of data records, the second artificial intelligence matching operation being based on the Dedupe Python library.

16. The system of claim 9, wherein the AI ​​matching operation based on the transformer model is performed on cloud infrastructure.

Images