Hugging Face's logo

heado
/
asr_mind_mode

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Model card Files
xet
 Metrics Community
heado commited on
Commit
ec010d1
·
verified ·
1 Parent(s): 25c3183

End of training

Browse files
Files changed (5) hide show
  1. README.md +12 -16
  2. config.json +2 -1
  3. model.safetensors +2 -2
  4. preprocessor_config.json +0 -1
  5. training_args.bin +2 -2
README.md CHANGED
@@ -4,8 +4,6 @@ license: apache-2.0
4
  base_model: w11wo/wav2vec2-xls-r-300m-korean
5
  tags:
6
  - generated_from_trainer
7
- metrics:
8
- - wer
9
  model-index:
10
  - name: asr_mind_mode
11
  results: []
@@ -18,8 +16,13 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [w11wo/wav2vec2-xls-r-300m-korean](https://huggingface.co/w11wo/wav2vec2-xls-r-300m-korean) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 2.7565
22
- - Wer: 4.3843
 
 
 
 
 
23
 
24
  ## Model description
25
 
@@ -47,19 +50,12 @@ The following hyperparameters were used during training:
47
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: linear
49
  - lr_scheduler_warmup_steps: 500
50
- - training_steps: 5
51
  - mixed_precision_training: Native AMP
52
 
53
- ### Training results
54
-
55
- | Training Loss | Epoch | Step | Validation Loss | Wer |
56
- |:-------------:|:------:|:----:|:---------------:|:------:|
57
- | 3.8183 | 0.1667 | 5 | 2.7565 | 4.3843 |
58
-
59
-
60
  ### Framework versions
61
 
62
- - Transformers 4.57.3
63
- - Pytorch 2.9.0+cu126
64
- - Datasets 4.0.0
65
- - Tokenizers 0.22.1
 
4
  base_model: w11wo/wav2vec2-xls-r-300m-korean
5
  tags:
6
  - generated_from_trainer
 
 
7
  model-index:
8
  - name: asr_mind_mode
9
  results: []
 
16
 
17
  This model is a fine-tuned version of [w11wo/wav2vec2-xls-r-300m-korean](https://huggingface.co/w11wo/wav2vec2-xls-r-300m-korean) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - eval_loss: 2.1909
20
+ - eval_wer: 0.7319
21
+ - eval_runtime: 15.7729
22
+ - eval_samples_per_second: 7.545
23
+ - eval_steps_per_second: 0.951
24
+ - epoch: 2.6667
25
+ - step: 80
26
 
27
  ## Model description
28
 
 
50
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: linear
52
  - lr_scheduler_warmup_steps: 500
53
+ - training_steps: 2000
54
  - mixed_precision_training: Native AMP
55
 
 
 
 
 
 
 
 
56
  ### Framework versions
57
 
58
+ - Transformers 5.20.1
59
+ - Pytorch 2.11.0+cu130
60
+ - Datasets 4.8.5
61
+ - Tokenizers 0.23.3
config.json CHANGED
@@ -100,7 +100,8 @@
100
  1,
101
  1
102
  ],
103
- "transformers_version": "4.57.3",
 
104
  "use_weighted_layer_sum": false,
105
  "vocab_size": 1207,
106
  "xvector_output_dim": 512
 
100
  1,
101
  1
102
  ],
103
+ "transformers_version": "5.20.1",
104
+ "use_cache": false,
105
  "use_weighted_layer_sum": false,
106
  "vocab_size": 1207,
107
  "xvector_output_dim": 512
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:54908828025406effe5217c255cfc7809332adea9f02d42278b54adefb2206a9
3
- size 1266756196
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a94db05309628bf54e66a5aa2d7b2ad66468163d13a4c8bdaf6fde5100b39364
3
+ size 1266756148
preprocessor_config.json CHANGED
@@ -4,7 +4,6 @@
4
  "feature_size": 1,
5
  "padding_side": "right",
6
  "padding_value": 0,
7
- "processor_class": "Wav2Vec2Processor",
8
  "return_attention_mask": true,
9
  "sampling_rate": 16000
10
  }
 
4
  "feature_size": 1,
5
  "padding_side": "right",
6
  "padding_value": 0,
 
7
  "return_attention_mask": true,
8
  "sampling_rate": 16000
9
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fb977019bf63cf8adcbf8c49f474e3a782eb512b767f8fd47854d549557faf84
3
- size 5841
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:738653ab19799bf5bc704e46311aa9241461cc38a0f0bfebf5f730e262ef956d
3
+ size 5201