Legal document review · Production

Love Without Sound builds NLP tools for music royalty recovery and legal correspondence processing

The problem

Music is frequently used without proper licensing across fragmented platforms, resulting in millions of lost royalties for artists who are powerless to monitor or claim compensation. Incorrect metadata compounds the problem — $2.5 billion in royalties remained unallocated in the U.S. alone between 2016 and 2018. Legal correspondence to address these violations is voluminous, with thousands of emails sent per day.

Workflow diagram · grounded in source
1
Royalty recovery case initiated
trigger
“Love Without Sound helps their labels find and recover royalties for music used without the appropriate license”
2
Metadata normalization pipeline
ai_action
“Jordan developed a spaCy pipeline with named entity recognition and text classification components that normalize and standardize song and artist information across a 2 billion-row database. The models extract components like song titles…”
3
Email content classification
ai_action
“the correspondence classifier distinguishes substantive business communications from non-essential emails like newsletters and meeting invitations”
4
Case citation and counter-argument recommendation
ai_action
“the system is able to recommend appropriate counter-arguments and predict the direction a case is heading in, given the arguments used and a large volume of previous negotiations”
5
Attachment and document extraction
ai_action
“In addition to the email contents, the attachments are classified as well and critical data points are extracted. For settlement agreements, the system identifies payment deadlines, amounts, and granted rights.”
6
Music reference linking
ai_action
“the system links these references to unique identifiers in its music metadata database”
7
Request tracking dashboard
output
“extract explicitly stated or implied action items and requests, classify their urgency and create a real-time dashboard of pending requests”
8
Continuous model retraining
feedback_loop
“If there are new client requests or additions to the music catalog, I can simply spin up Prodigy, create a new dataset with more examples and edge cases, and update the model. This keeps the results current and gives me a consistent and …”
Reported outcome

Love Without Sound's tools helped publishers recover hundreds of millions of dollars in lost revenue for artists and reduced legal research time by nearly 50%.
Models run in a data-private environment and handle real-time processing at scale.

Reported metrics
Royalties recovered for artistshundreds of millions of dollars
Legal research time reductionnearly 50%
unallocated royalties U.S. 2016–2018 (problem context)$2.5 billion
Reported stack
spaCyProdigyModalRetrieval-Augmented Generation (RAG)
Source
https://explosion.ai/blog/love-without-sound-nlp-music-industry
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Love Without Sound's tools helped publishers recover hundreds of millions of dollars in lost revenue for artists and reduced legal research time by nearly 50%.

What tools did this team use?

spaCy, Prodigy, Modal, Retrieval-Augmented Generation (RAG).

What results were reported?

Royalties recovered for artists: hundreds of millions of dollars; Legal research time reduction: nearly 50%; unallocated royalties U.S. 2016–2018 (problem context): $2.5 billion (source-reported, not independently verified).

How is this legal document review AI workflow structured?

Royalty recovery case initiated → Metadata normalization pipeline → Email content classification → Case citation and counter-argument recommendation → Attachment and document extraction → Music reference linking → Request tracking dashboard → Continuous model retraining.

Related legal document review cases