Hugging Face's logo

Akashpb13
/
Hausa_xlsr

Automatic Speech Recognition
Transformers
PyTorch
Hausa
wav2vec2
Generated from Trainer
hf-asr-leaderboard
model_for_talk
mozilla-foundation/common_voice_8_0
robust-speech-event
Eval Results
Inference Endpoints
Model card Files Community
2
Akashpb13 commited on
Commit
9deb9d5
•
1 Parent(s): b653a65

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -20
README.md CHANGED
@@ -21,28 +21,28 @@ model-index:
21
  dataset:
22
  name: Common Voice 8
23
  type: mozilla-foundation/common_voice_8_0
24
- args: kmr
25
  metrics:
26
  - name: Test WER
27
  type: wer
28
- value: 0.21177149451817656
29
  - name: Test CER
30
  type: cer
31
- value: 0.04472733528298641
32
  - task:
33
  name: Automatic Speech Recognition
34
  type: automatic-speech-recognition
35
  dataset:
36
  name: Robust Speech Event - Dev Data
37
  type: speech-recognition-community-v2/dev_data
38
- args: kmr
39
  metrics:
40
  - name: Test WER
41
  type: wer
42
- value: 0.21177149451817656
43
  - name: Test CER
44
  type: cer
45
- value: 0.04472733528298641
46
  ---
47
 
48
  # Akashpb13/xlsr_hungarian_new
@@ -83,26 +83,26 @@ The following hyperparameters were used during training:
83
 
84
  | Step | Training Loss | Validation Loss | Wer |
85
  |------|---------------|-----------------|----------|
86
- | 500 | 5.176500 | 2.776366 | 1.000000 |
87
- | 1000 | 1.166500 | 0.329668 | 0.482262 |
88
- | 1500 | 0.334500 | 0.245564 | 0.391304 |
89
- | 2000 | 0.226300 | 0.241998 | 0.381168 |
90
- | 2500 | 0.174600 | 0.227778 | 0.356895 |
91
- | 3000 | 0.141700 | 0.237551 | 0.343292 |
92
- | 3500 | 0.117600 | 0.259884 | 0.348626 |
93
- | 4000 | 0.097800 | 0.261468 | 0.343825 |
94
- | 4500 | 0.083500 | 0.268763 | 0.332889 |
95
- | 5000 | 0.074500 | 0.276381 | 0.332622 |
96
- | 5500 | 0.067700 | 0.275504 | 0.328354 |
97
- | 6000 | 0.064500 | 0.274393 | 0.331288 |
98
- | 6500 | 0.061400 | 0.275118 | 0.329955 |
99
 
100
 
101
 
102
  ### Framework versions
103
  - Transformers 4.16.0.dev0
104
  - Pytorch 1.10.0+cu102
105
- - Datasets 1.17.1.dev0
106
  - Tokenizers 0.10.3
107
 
108
  #### Evaluation Commands
 
21
  dataset:
22
  name: Common Voice 8
23
  type: mozilla-foundation/common_voice_8_0
24
+ args: ha
25
  metrics:
26
  - name: Test WER
27
  type: wer
28
+ value: 0.20614541257934219
29
  - name: Test CER
30
  type: cer
31
+ value: 0.04358048053214061
32
  - task:
33
  name: Automatic Speech Recognition
34
  type: automatic-speech-recognition
35
  dataset:
36
  name: Robust Speech Event - Dev Data
37
  type: speech-recognition-community-v2/dev_data
38
+ args: ha
39
  metrics:
40
  - name: Test WER
41
  type: wer
42
+ value: 0.20614541257934219
43
  - name: Test CER
44
  type: cer
45
+ value: 0.04358048053214061
46
  ---
47
 
48
  # Akashpb13/xlsr_hungarian_new
 
83
 
84
  | Step | Training Loss | Validation Loss | Wer |
85
  |------|---------------|-----------------|----------|
86
+ | 500 | 5.175900 | 2.750914 | 1.000000 |
87
+ | 1000 | 1.028700 | 0.338649 | 0.497999 |
88
+ | 1500 | 0.332200 | 0.246896 | 0.402241 |
89
+ | 2000 | 0.227300 | 0.239640 | 0.395839 |
90
+ | 2500 | 0.175000 | 0.239577 | 0.373966 |
91
+ | 3000 | 0.140400 | 0.243272 | 0.356095 |
92
+ | 3500 | 0.119200 | 0.263761 | 0.365164 |
93
+ | 4000 | 0.099300 | 0.265954 | 0.353428 |
94
+ | 4500 | 0.084400 | 0.276367 | 0.349693 |
95
+ | 5000 | 0.073700 | 0.282631 | 0.343825 |
96
+ | 5500 | 0.068000 | 0.282344 | 0.341158 |
97
+ | 6000 | 0.064500 | 0.281591 | 0.342491 |
98
+
99
 
100
 
101
 
102
  ### Framework versions
103
  - Transformers 4.16.0.dev0
104
  - Pytorch 1.10.0+cu102
105
+ - Datasets 1.18.3
106
  - Tokenizers 0.10.3
107
 
108
  #### Evaluation Commands