| audio: | |
| chunk_size: 131584 | |
| dim_f: 1024 | |
| dim_t: 256 | |
| hop_length: 512 | |
| n_fft: 2048 | |
| num_channels: 2 | |
| sample_rate: 44100 | |
| min_mean_abs: 0.001 | |
| model: | |
| dim: 192 | |
| depth: 8 | |
| stereo: true | |
| num_stems: 1 | |
| time_conformer_depth: 1 | |
| freq_conformer_depth: 1 | |
| # band-splitting | |
| num_bands: 60 | |
| # attention/width | |
| dim_head: 64 | |
| heads: 8 | |
| ff_mult: 4 | |
| # conformer conv sub-block | |
| conv_expansion_factor: 2 | |
| conv_kernel_size: 31 | |
| # dropouts | |
| attn_dropout: 0.0 | |
| ff_dropout: 0.0 | |
| conv_dropout: 0.0 | |
| # STFT | |
| dim_freqs_in: 1025 | |
| sample_rate: 44100 | |
| stft_n_fft: 2048 | |
| stft_hop_length: 512 | |
| stft_win_length: 2048 | |
| stft_normalized: False | |
| # mask estimator + multi-res STFT loss | |
| mask_estimator_depth: 2 | |
| multi_stft_resolution_loss_weight: 1.0 | |
| multi_stft_resolutions_window_sizes: !!python/tuple | |
| - 4096 | |
| - 2048 | |
| - 1024 | |
| - 512 | |
| - 256 | |
| multi_stft_hop_size: 147 | |
| multi_stft_normalized: False | |
| use_torch_checkpoint: False | |
| skip_connection: False | |
| match_input_audio_length: False | |
| training: | |
| batch_size: 1 | |
| gradient_accumulation_steps: 1 | |
| grad_clip: 0 | |
| instruments: | |
| - vocals | |
| - bass | |
| - drums | |
| - other | |
| lr: 5.0e-05 | |
| patience: 2 | |
| reduce_factor: 0.95 | |
| target_instrument: vocals | |
| num_epochs: 1000 | |
| num_steps: 1000 | |
| q: 0.95 | |
| coarse_loss_clip: true | |
| ema_momentum: 0.999 | |
| optimizer: adam | |
| other_fix: false | |
| use_amp: true | |
| augmentations: | |
| enable: false | |
| loudness: true | |
| loudness_min: 0.5 | |
| loudness_max: 1.5 | |
| inference: | |
| batch_size: 2 | |
| dim_t: 256 | |
| num_overlap: 2 | |
| chunk_size: 131584 | |