Hugging Face's logo

hynt
/
EfficientConformerVietnamese

PyTorch
Vietnamese
CTC
speech-to-text
vietnamese
ai-model
deep-learning
Model card Files
xet
Community
hynt commited on
Commit
3904444
·
verified ·
1 Parent(s): c084131

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -64,10 +64,10 @@ tensorboard --logdir callback_path
64
 
65
  ## Vietnamese Performance
66
 
67
- | Model | Gigaspeech_test<br>(Greedy / n-gram Beam Search) | VLSP2023_pb_test<br>(Greedy / n-gram Beam Search) | VLSP2023_pr_test<br>(Greedy / n-gram Beam Search) |
68
- |:-----------------------------|:------------------------------------------------:|:-------------------------------------------------:|:-------------------------------------------------:|
69
- | Efficient-Conformer-Small-CTC| **19.61 / 17.86** | **23.06 / 21.11** | **23.17 / 21.42** |
70
 
 
 
 
71
 
72
 
73
  In the competition organized by VLSP, I used the Efficient Conformer Large architecture with approximately 127 million parameters. You can find the detailed results in the technical report below:
 
64
 
65
  ## Vietnamese Performance
66
 
 
 
 
67
 
68
+ | Model | Gigaspeech_test<br>(Greedy / n-gram Beam Search) | VLSP2023_pb_test<br>(Greedy / n-gram Beam Search) | VLSP2023_pr_test<br>(Greedy / n-gram Beam Search) |
69
+ |:--------------------------------------|:------------------------------------------------:|:-------------------------------------------------:|:-------------------------------------------------:|
70
+ | **Efficient-Conformer-Small-CTC** | **19.61 / 17.86** | **23.06 / 21.11** | **23.17 / 21.42** |
71
 
72
 
73
  In the competition organized by VLSP, I used the Efficient Conformer Large architecture with approximately 127 million parameters. You can find the detailed results in the technical report below: