Initial commit
Browse files- README.md +82 -0
- pytorch_model.bin +3 -0
README.md
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- asteroid
|
4 |
+
- audio
|
5 |
+
- ConvTasNet
|
6 |
+
datasets:
|
7 |
+
- Libri1Mix
|
8 |
+
- enh_single
|
9 |
+
license: CC BY-NC 4.0
|
10 |
+
inference: false
|
11 |
+
---
|
12 |
+
|
13 |
+
## Asteroid model `JorisCos/DPRNNTasNet_Libri1Mix_enhsignle_16k`
|
14 |
+
|
15 |
+
Description:
|
16 |
+
|
17 |
+
This model was trained by Joris Cosentino using the librimix recipe in [Asteroid](https://github.com/asteroid-team/asteroid).
|
18 |
+
It was trained on the `enh_single` task of the Libri1Mix dataset.
|
19 |
+
|
20 |
+
Training config:
|
21 |
+
|
22 |
+
```yml
|
23 |
+
data:
|
24 |
+
n_src: 1
|
25 |
+
sample_rate: 16000
|
26 |
+
segment: 1
|
27 |
+
task: enh_single
|
28 |
+
train_dir: data/wav16k/min/train-360
|
29 |
+
valid_dir: data/wav16k/min/dev
|
30 |
+
filterbank:
|
31 |
+
kernel_size: 2
|
32 |
+
n_filters: 64
|
33 |
+
stride: 1
|
34 |
+
masknet:
|
35 |
+
bidirectional: true
|
36 |
+
bn_chan: 128
|
37 |
+
chunk_size: 250
|
38 |
+
dropout: 0
|
39 |
+
hid_size: 128
|
40 |
+
hop_size: 125
|
41 |
+
in_chan: 64
|
42 |
+
mask_act: sigmoid
|
43 |
+
n_repeats: 6
|
44 |
+
n_src: 1
|
45 |
+
out_chan: 64
|
46 |
+
optim:
|
47 |
+
lr: 0.001
|
48 |
+
optimizer: adam
|
49 |
+
weight_decay: 1.0e-05
|
50 |
+
training:
|
51 |
+
batch_size: 2
|
52 |
+
early_stop: true
|
53 |
+
epochs: 200
|
54 |
+
gradient_clipping: 5
|
55 |
+
half_lr: true
|
56 |
+
num_workers: 4
|
57 |
+
```
|
58 |
+
|
59 |
+
|
60 |
+
Results:
|
61 |
+
|
62 |
+
On Libri1Mix min test set :
|
63 |
+
```yml
|
64 |
+
si_sdr: 14.7228101708889,
|
65 |
+
si_sdr_imp: 11.2730288650292,
|
66 |
+
sdr: 15.35661405197161,
|
67 |
+
sdr_imp: 11.853951252758595,
|
68 |
+
sir: Infinity,
|
69 |
+
sir_imp: NaN,
|
70 |
+
sar: 15.35661405197161,
|
71 |
+
sar_imp: 11.853951252758595,
|
72 |
+
stoi: 0.9300461826351578,
|
73 |
+
stoi_imp: 0.13412635909461715
|
74 |
+
```
|
75 |
+
|
76 |
+
|
77 |
+
License notice:
|
78 |
+
|
79 |
+
This work "DPRNNTasNet_Libri1Mix_enhsignle_16k" is a derivative of [LibriSpeech ASR corpus](http://www.openslr.org/12) by Vassil Panayotov,
|
80 |
+
used under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/); of The WSJ0 Hipster Ambient Mixtures
|
81 |
+
dataset by [Whisper.ai](http://wham.whisper.ai/), used under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/) (Research only).
|
82 |
+
"DPRNNTasNet_Libri1Mix_enhsignle_16k" is licensed under [Attribution-ShareAlike 3.0 Unported](https://creativecommons.org/licenses/by-sa/3.0/) by Joris Cosentino
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6b510c07fae3a3db18473b5749316cb9df8dc4f78164c3cdfbb50d3783ee779d
|
3 |
+
size 14595773
|