3 ms·
It's lossy, the inverse is not well defined. But you could try to train a NN on reconstructing plausible (idiomatic) inputs leading to the current IR...
by Certhas 6y ago
It's lossy, the inverse is not well defined. But you could try to train a NN on reconstructing plausible (idiomatic) inputs leading to the current IR...