tags:
- asteroid
- audio
- MultiDecoderDPRNN
datasets:
- Wsj0MixVar
- sep_clean
inference: false
Asteroid model
Description:
Code: The code corresponding to this model card can be found in the astroid toolkit @ https://github.com/asteroid-team/asteroid, under "egs/wsj0-mix-var", where the recipe is stored.
Paper: "Multi-Decoder DPRNN: High Accuracy Source Counting and Separation", Junzhe Zhu, Raymond Yeh, Mark Hasegawa-Johnson. ICASSP(2021). https://ieeexplore.ieee.org/document/9414205
Summary: This model achieves SOTA on the problem of source separation with an unknown number of speakers. It uses multiple decoder heads(each tackling a distinct number of speakers), in addition to a classifier head that selects which decoder head to use.
Demo Page: https://junzhejosephzhu.github.io/Multi-Decoder-DPRNN/
Original research repo is at https://github.com/JunzheJosephZhu/MultiDecoder-DPRNN
This model was trained by Joseph Zhu using the wsj0-mix-var/Multi-Decoder-DPRNN recipe in Asteroid.
It was trained on the sep_count task of the Wsj0MixVar dataset.
Training config:
filterbank:
n_filters: 64
kernel_size: 8
stride: 4
masknet:
n_srcs: [2, 3, 4, 5]
bn_chan: 128
hid_size: 128
chunk_size: 128
hop_size: 64
n_repeats: 8
mask_act: 'sigmoid'
bidirectional: true
dropout: 0
use_mulcat: false
training:
epochs: 200
batch_size: 2
num_workers: 2
half_lr: yes
lr_decay: yes
early_stop: yes
gradient_clipping: 5
optim:
optimizer: adam
lr: 0.001
weight_decay: 0.00000
data:
train_dir: "data/{}speakers/wav8k/min/tr"
valid_dir: "data/{}speakers/wav8k/min/cv"
task: sep_count
sample_rate: 8000
seglen: 4.0
minlen: 2.0
loss:
lambda: 0.05
Results:
'Accuracy': 0.9723333333333334, 'P-Si-SNR': 10.36027378628496