-
card: encoder ladder (3 window lengths), new default numbers
-
Encoder_sparse-g8.mlmodelc: the previous single-shape 15 s default, kept for reference
-
Encoder.mlmodelc: multifunction ladder (enc_5040ms / enc_10160ms / enc_15000ms over one weight blob)
-
card: FluidAudio #980 merged
-
card: describe the encoding instead of linking the recipe
-
card: same-session v3/Ultra/Phonon-2 LibriSpeech table
-
card: 60-minute long-form comparison
-
Phonon-2 Core ML: 5 exact five-value encoders (sparse-g8 default), decoder, joint, preprocessor, vocab, attribution
-
initial commit
