Overview
What it is
Audioneex is a high-performance Audio Content Recognition system (ACR) based on "audio fingerprinting" technology designed for real-time applications and resource-constrained devices. It's particularly suited for native deployments on mobile and embedded systems but can be used on pretty much any machine.
Working principles
The engine was initially developed for identification of music recordings, but its core design is based on machine learning algorithms that can be easily used to retrain the stock model for specific use cases. The stock model, however, is generic enough for uses as a general-purpose ACR system on different kinds of audio. Note, however, that despite using ML techniques, Audioneex is an audio fingerprinting system designed to recognize predetermined instances of audio signals and not a classifier.
Common use cases
ACR systems can be used in a variety of scenarios, such as broadcast monitoring, content synchronization, second screen, audio surveillance, etc. Audio content identification and management technology finds applications in several industries. Audioneex provides the core technology for such applications in the form of a C++ cross-platform API that can be integrated as a backend component in web services, mobile and desktop apps, and embedded systems. For more detailed information and the full API specifications, refer to the documentation.
