Code moved to compute_frame_features()

Starting with compute_linear()

Remove already-disabled silence chopping code and make time alignment
the same as lpcnet_demo -features

We're back to processing 10 ms at a time and have no need for 40-ms
"superframes".

Krishna Subramani authored 3 years ago and Jean-Marc Valin committed 3 years ago

Making LPC computation and prediction differentiable

Jean-Marc Valin authored 3 years ago and Jean-Marc Valin committed 3 years ago

Down to 20 features

was increased too much in 713d53e8a