Our in house ML models can estimate pitch and extract chords from audio streams, on the fly, in realtime. Our proprietary models can estimate the 3d position and orientation of real instruments from a single photograph. The algorithms are trained using a mix of real and synthetic data, and can work with reflective surfaces and repeating patterns. We've developed new machine learning algorithms that can synthesize novel and kinematically accurate 3d musical performance from just a midi audio file, for the use in education and AR / VR. Our tools can perform advanced full body and hand inverse kinematics to fit the same 3d musical performance to different avatars.