Hugging Face's logo

khanhld
/
chunkformer-large-en-libri-960h

Automatic Speech Recognition
PyTorch
English
transcription
audio
speech
chunkformer
asr
long-form transcription
librispeech
Eval Results
Model card Files Community
khanhld commited on
Commit
9cee4fc
·
1 Parent(s): 4716077

[update] update model checkpoint, config, results

Browse files
Files changed (3) hide show
  1. README.md +1 -1
  2. config.yaml +1 -1
  3. pytorch_model.bin +2 -2
README.md CHANGED
@@ -74,7 +74,7 @@ We evaluate the models using **Word Error Rate (WER)**. To ensure a fair compari
74
 
75
  | STT | Model | Test-Clean | Test-Other | Avg. |
76
  |-----|-----------------------|------------|------------|------ |
77
- | 1 | **ChunkFormer** | 2.69 | 6.89 | 4.79 |
78
  | 2 | **Efficient Conformer** | 2.71 | 6.95 | 4.83 |
79
  | 3 | **Conformer** | 2.77 | 6.93 | 4.85 |
80
  | 4 | **Squeezeformer** | 2.87 | 7.16 | 5.02 |
 
74
 
75
  | STT | Model | Test-Clean | Test-Other | Avg. |
76
  |-----|-----------------------|------------|------------|------ |
77
+ | 1 | **ChunkFormer** | 2.69 | 6.91 | 4.80 |
78
  | 2 | **Efficient Conformer** | 2.71 | 6.95 | 4.83 |
79
  | 3 | **Conformer** | 2.77 | 6.93 | 4.85 |
80
  | 4 | **Squeezeformer** | 2.87 | 7.16 | 5.02 |
config.yaml CHANGED
@@ -1,6 +1,6 @@
1
  encoder: chunkformer
2
  is_json_cmvn: true
3
- cmvn_file: chunkformer-large-en/global_cmvn
4
  input_dim: 80
5
  output_dim: 4990
6
 
 
1
  encoder: chunkformer
2
  is_json_cmvn: true
3
+ cmvn_file: chunkformer-large-en-libri-960h/global_cmvn
4
  input_dim: 80
5
  output_dim: 4990
6
 
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:015c879bbae04ef50b668bf6b54ae0777c9b461a2fa4db9ab485ef555d10cb04
3
- size 593420152
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:568924c7472dfd7128f16088ef4fdb94296728a50ba7879a7fda8aea685c3ab3
3
+ size 593426663