Hugging Face's logo

ctaguchi
/
ssc-sco-mms-model

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Model card Files
xet
 Community
ctaguchi commited on
Commit
72e87c0
·
verified ·
1 Parent(s): 475fc2c

Model save

Browse files
Files changed (2) hide show
  1. README.md +44 -8
  2. model.safetensors +1 -1
README.md CHANGED
@@ -18,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.8062
22
- - Cer: 0.2216
23
- - Wer: 0.5801
24
 
25
  ## Model description
26
 
@@ -48,17 +48,53 @@ The following hyperparameters were used during training:
48
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
49
  - lr_scheduler_type: linear
50
  - lr_scheduler_warmup_steps: 100
51
- - num_epochs: 1
52
  - mixed_precision_training: Native AMP
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Cer | Wer |
57
  |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
58
- | 2.0546 | 0.2491 | 200 | 0.9257 | 0.2435 | 0.6046 |
59
- | 1.9143 | 0.4981 | 400 | 0.8285 | 0.2265 | 0.6000 |
60
- | 1.8565 | 0.7472 | 600 | 0.8162 | 0.2235 | 0.5908 |
61
- | 1.7833 | 0.9963 | 800 | 0.8062 | 0.2216 | 0.5801 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
62
 
63
 
64
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.7277
22
+ - Cer: 0.2065
23
+ - Wer: 0.5486
24
 
25
  ## Model description
26
 
 
48
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
49
  - lr_scheduler_type: linear
50
  - lr_scheduler_warmup_steps: 100
51
+ - num_epochs: 10
52
  - mixed_precision_training: Native AMP
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Cer | Wer |
57
  |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
58
+ | 2.0529 | 0.2491 | 200 | 0.9213 | 0.2438 | 0.6057 |
59
+ | 1.9092 | 0.4981 | 400 | 0.8217 | 0.2253 | 0.5961 |
60
+ | 1.85 | 0.7472 | 600 | 0.7991 | 0.2205 | 0.5779 |
61
+ | 1.7687 | 0.9963 | 800 | 0.7940 | 0.2199 | 0.5727 |
62
+ | 1.8151 | 1.2453 | 1000 | 0.7970 | 0.2182 | 0.5831 |
63
+ | 1.7891 | 1.4944 | 1200 | 0.7940 | 0.2177 | 0.5835 |
64
+ | 1.7512 | 1.7435 | 1400 | 0.7717 | 0.2149 | 0.5776 |
65
+ | 1.7742 | 1.9925 | 1600 | 0.7779 | 0.2149 | 0.5780 |
66
+ | 1.7784 | 2.2416 | 1800 | 0.7787 | 0.2140 | 0.5807 |
67
+ | 1.7212 | 2.4907 | 2000 | 0.7709 | 0.2133 | 0.5745 |
68
+ | 1.7672 | 2.7397 | 2200 | 0.7663 | 0.2138 | 0.5739 |
69
+ | 1.7632 | 2.9888 | 2400 | 0.7594 | 0.2104 | 0.5667 |
70
+ | 1.7459 | 3.2379 | 2600 | 0.7738 | 0.2125 | 0.5743 |
71
+ | 1.7162 | 3.4869 | 2800 | 0.7740 | 0.2114 | 0.5780 |
72
+ | 1.7718 | 3.7360 | 3000 | 0.7873 | 0.2145 | 0.5867 |
73
+ | 1.7002 | 3.9851 | 3200 | 0.7640 | 0.2104 | 0.5758 |
74
+ | 1.7167 | 4.2341 | 3400 | 0.7606 | 0.2121 | 0.5738 |
75
+ | 1.7429 | 4.4832 | 3600 | 0.7545 | 0.2100 | 0.5653 |
76
+ | 1.7338 | 4.7323 | 3800 | 0.7525 | 0.2092 | 0.5639 |
77
+ | 1.7302 | 4.9813 | 4000 | 0.7529 | 0.2088 | 0.5615 |
78
+ | 1.6996 | 5.2304 | 4200 | 0.7480 | 0.2090 | 0.5634 |
79
+ | 1.7655 | 5.4795 | 4400 | 0.7445 | 0.2083 | 0.5649 |
80
+ | 1.7414 | 5.7285 | 4600 | 0.7449 | 0.2091 | 0.5617 |
81
+ | 1.6881 | 5.9776 | 4800 | 0.7520 | 0.2107 | 0.5675 |
82
+ | 1.7261 | 6.2267 | 5000 | 0.7571 | 0.2097 | 0.5730 |
83
+ | 1.7112 | 6.4757 | 5200 | 0.7438 | 0.2072 | 0.5601 |
84
+ | 1.7348 | 6.7248 | 5400 | 0.7438 | 0.2068 | 0.5573 |
85
+ | 1.6992 | 6.9738 | 5600 | 0.7372 | 0.2068 | 0.5573 |
86
+ | 1.7178 | 7.2229 | 5800 | 0.7350 | 0.2066 | 0.5545 |
87
+ | 1.7061 | 7.4720 | 6000 | 0.7332 | 0.2072 | 0.5554 |
88
+ | 1.6895 | 7.7210 | 6200 | 0.7353 | 0.2079 | 0.5571 |
89
+ | 1.7274 | 7.9701 | 6400 | 0.7317 | 0.2072 | 0.5534 |
90
+ | 1.7295 | 8.2192 | 6600 | 0.7314 | 0.2068 | 0.5517 |
91
+ | 1.6925 | 8.4682 | 6800 | 0.7308 | 0.2068 | 0.5525 |
92
+ | 1.7261 | 8.7173 | 7000 | 0.7280 | 0.2064 | 0.5485 |
93
+ | 1.762 | 8.9664 | 7200 | 0.7271 | 0.2067 | 0.5466 |
94
+ | 1.6809 | 9.2154 | 7400 | 0.7276 | 0.2068 | 0.5477 |
95
+ | 1.7149 | 9.4645 | 7600 | 0.7271 | 0.2066 | 0.5485 |
96
+ | 1.6764 | 9.7136 | 7800 | 0.7276 | 0.2067 | 0.5488 |
97
+ | 1.6877 | 9.9626 | 8000 | 0.7277 | 0.2065 | 0.5486 |
98
 
99
 
100
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:53b0d824bccaacb588262914ac1b41ee0e6d40587030e6b3e24e4c17bd693560
3
  size 3859521176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cf6695d0e8201c38292370f4754362e11b4dc215083380d2e6e1dd1cbf95ea3
3
  size 3859521176