Audio Data Recognition

What is it?

In the context of libraries, audio data recognition is currently used more for the purposes and benefit of researchers (Stover, 2021).

Music information retrieval is a type of audio data recognition that is more advanced and seeks to improve existing audio retrieval frameworks for audio analysis across complex, large collections (Stover, 2021). Audio retrieval frameworks and infrastructure that already exist include tools such as algorithms that allow searching, analysis, encoding, and transformation of data associated with allowing the researcher to retrieve and access the data for specific resources in a collection (Stover, 2021).

Specifically, music libraries can benefit from the retrieval, analysis, and linking of audio metadata between resources that audio data recognition services can offer to users who are comparing music on a wide variety of fronts. Audio data recognition could be used to assist patrons by listening to their questions and redirecting them to the resources they are searching for. Systems that use machine learning to listen to audio for analysis, however, do cause valid privacy concerns because it is not known by the average person when these devices will be listening or not.