Hugging Face's logo

nvidia
/
stt_ua_fastconformer_hybrid_large_pc

Automatic Speech Recognition
NeMo
PyTorch
Ukrainian
speech
audio
Transducer
FastConformer
CTC
Transformer
NeMo
hf-asr-leaderboard
Eval Results (legacy)
Model card Files
xet
 Community
4

Update README.md

#3
by nkarpov - opened
base: refs/heads/main
←
from: refs/pr/3
Discussion Files changed
+4
-4
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -34,7 +34,7 @@ model-index:
34
  metrics:
35
  - name: Test WER
36
  type: wer
37
- value: 5.53
38
  - task:
39
  name: Automatic Speech Recognition
40
  type: automatic-speech-recognition
@@ -48,7 +48,7 @@ model-index:
48
  metrics:
49
  - name: Test WER P&C
50
  type: wer
51
- value: 6.92
52
  ---
53
 
54
  # NVIDIA FastConformer-Hybrid Large (ua)
@@ -148,13 +148,13 @@ The following tables summarizes the performance of the available models in this
148
  a) On data without Punctuation and Capitalization with Transducer decoder
149
  | **Version** | **Tokenizer** | **Vocabulary Size** | **MCV12 TEST** |
150
  |:-----------:|:---------------------:|:-------------------:|:-------------:|
151
- | 1.0.0 | SentencePiece Unigram | 512 | 5.53 |
152
 
153
 
154
  b) On data with Punctuation and Capitalization with Transducer decoder
155
  | **Version** | **Tokenizer** | **Vocabulary Size** | **MCV12 TEST** |
156
  |:-----------:|:---------------------:|:-------------------:|:-------------:|
157
- | 1.0.0 | SentencePiece Unigram | 512 | 6.92 |
158
 
159
 
160
  ## Limitations
 
34
  metrics:
35
  - name: Test WER
36
  type: wer
37
+ value: 5.66
38
  - task:
39
  name: Automatic Speech Recognition
40
  type: automatic-speech-recognition
 
48
  metrics:
49
  - name: Test WER P&C
50
  type: wer
51
+ value: 8.06
52
  ---
53
 
54
  # NVIDIA FastConformer-Hybrid Large (ua)
 
148
  a) On data without Punctuation and Capitalization with Transducer decoder
149
  | **Version** | **Tokenizer** | **Vocabulary Size** | **MCV12 TEST** |
150
  |:-----------:|:---------------------:|:-------------------:|:-------------:|
151
+ | 1.0.2 | SentencePiece Unigram | 512 | 5.66 |
152
 
153
 
154
  b) On data with Punctuation and Capitalization with Transducer decoder
155
  | **Version** | **Tokenizer** | **Vocabulary Size** | **MCV12 TEST** |
156
  |:-----------:|:---------------------:|:-------------------:|:-------------:|
157
+ | 1.0.2 | SentencePiece Unigram | 512 | 8.06 |
158
 
159
 
160
  ## Limitations