模型:
mpariente/ConvTasNet_Libri3Mix_sepnoisy
任务:
音频到音频许可:
cc-by-sa-4.0从此Zenodo导入 model page 。
此模型由Takhir Mirzaev使用Asteroid中的Librimix/ConvTasNet配方进行训练。它是在Libri3Mix数据集的sep_noisy任务上进行训练的。
data: n_src: 3 sample_rate: 8000 segment: 3 task: sep_noisy train_dir: data/wav8k/min/train-360 valid_dir: data/wav8k/min/dev filterbank: kernel_size: 16 n_filters: 512 stride: 8 masknet: bn_chan: 128 hid_chan: 512 mask_act: relu n_blocks: 8 n_repeats: 3 skip_chan: 128 optim: lr: 0.001 optimizer: adam weight_decay: 0.0 positional arguments: training: batch_size: 4 early_stop: True epochs: 200 half_lr: True num_workers: 4
si_sdr: 6.824750632456865 si_sdr_imp: 11.234803761803752 sdr: 7.715799858488098 sdr_imp: 11.778681386239114 sir: 16.442141130818637 sir_imp: 19.527535070051055 sar: 8.757864265661263 sar_imp: -0.15657258049670303 stoi: 0.7854554136619554 stoi_imp: 0.22267957718163015
这个作品"ConvTasNet_Libri3Mix_sepnoisy"是 Vassil Panayotov 创作的 LibriSpeech ASR corpus 的派生作品,根据 CC BY 4.0 使用。"ConvTasNet_Libri3Mix_sepnoisy"由Manuel Pariente根据 Attribution-ShareAlike 3.0 Unported 许可。