Managing hundreds of thousands of audio files can get messy over time. Sometimes, speech, sound effects, and other types of audio get mixed up with the music that you want to focus on for analysis.
To streamline the process of tagging and reduce unnecessary errors, we developed an Audio Type classifier to ‘pre-sort’ audio files into various categories. This keeps your catalog organized and prevents contamination of your music data.
Let’s take a look at our categories!
Definition: A track consisting of a pattern of organic or synthesized sounds with perceptible intent, designed to give pleasure and communicate emotions and/or ideas to listeners.
Definition: Full-length tracks that collectively contain the musical constituents of a full piece of Music, grouped by an audible logic (instrument, timbre, frequency band etc).
Definition: A short musical segment (<60s) featuring one or more complete musical phrase(s).
Definition: A short musical segment (<20s) featuring one or more complete musical phrase(s).
Definition: Natural or synthesized sounds (other than speech or music)
Definition: An audio file that consists of no more than two transients.
Definition: An audio file that primarily consists of sustained sound(s). Some music may have drone-like qualities and fall into both categories.
Definition: A recording of intelligible speech by any number of individual entities, in existing or invented languages, with no perceptible musical intent (organization of pitch and/or rhythm).
Definition: Any audio file that does not fit into any of the above categories.
Our Type Classifier is available in our Tag, Search, and API solutions.