Hugging Face's logo

Zhaoheng
/
svoice_wsj0_2mix

Audio-to-Audio
ESPnet
English
audio
Model card Files Community
Zhaoheng Ni commited on
Commit
9017d0a
•
1 Parent(s): 686ba10

update si_snr result

Browse files
Files changed (1) hide show
  1. README.md +5 -5
README.md CHANGED
@@ -40,10 +40,10 @@ cd egs2/wsj0_2mix/enh1
40
 
41
  config: conf/tuning/train_enh_svoice.yaml
42
 
43
- |dataset|STOI|SAR|SDR|SIR|
44
- |---|---|---|---|---|
45
- |enhanced_cv_min_8k|0.97|21.44|20.98|32.21|
46
- |enhanced_tt_min_8k|0.98|21.41|20.96|32.27|
47
 
48
  ## ENH config
49
 
@@ -56,7 +56,7 @@ log_level: INFO
56
  dry_run: false
57
  iterator_type: chunk
58
  output_dir: exp/enh_train_enh_svoice_raw
59
- ngpu: 1
60
  seed: 0
61
  num_workers: 4
62
  num_att_plot: 3
 
40
 
41
  config: conf/tuning/train_enh_svoice.yaml
42
 
43
+ |dataset|STOI|SAR|SDR|SIR|SI_SNR|
44
+ |---|---|---|---|---|---|
45
+ |enhanced_cv_min_8k|0.97|21.44|20.98|32.21|20.67|
46
+ |enhanced_tt_min_8k|0.98|21.41|20.96|32.27|20.66|
47
 
48
  ## ENH config
49
 
 
56
  dry_run: false
57
  iterator_type: chunk
58
  output_dir: exp/enh_train_enh_svoice_raw
59
+ ngpu: 4
60
  seed: 0
61
  num_workers: 4
62
  num_att_plot: 3