zainulhakim commited on
Commit
e3010c3
1 Parent(s): 71e9741

240624-wav2vec2-ASR-Arab

Browse files
Files changed (1) hide show
  1. README.md +9 -9
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  license: apache-2.0
3
- base_model: zainulhakim/240624-wav2vec2-ASR-Arab
4
  tags:
5
  - generated_from_trainer
6
  metrics:
@@ -61,12 +61,13 @@ should probably proofread and complete it, then remove this comment. -->
61
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/projektarbeit_dalim/huggingface/runs/ogklm4ck)
62
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/projektarbeit_dalim/huggingface/runs/ogklm4ck)
63
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/projektarbeit_dalim/huggingface/runs/ogklm4ck)
 
64
  # 240624-wav2vec2-ASR-Arab
65
 
66
- This model is a fine-tuned version of [zainulhakim/240624-wav2vec2-ASR-Arab](https://huggingface.co/zainulhakim/240624-wav2vec2-ASR-Arab) on an unknown dataset.
67
  It achieves the following results on the evaluation set:
68
- - Loss: 1.2948
69
- - Wer: 0.4141
70
 
71
  ## Model description
72
 
@@ -92,17 +93,16 @@ The following hyperparameters were used during training:
92
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
93
  - lr_scheduler_type: linear
94
  - lr_scheduler_warmup_steps: 500
95
- - num_epochs: 30
96
  - mixed_precision_training: Native AMP
97
 
98
  ### Training results
99
 
100
  | Training Loss | Epoch | Step | Validation Loss | Wer |
101
  |:-------------:|:-----:|:----:|:---------------:|:------:|
102
- | No log | 6.25 | 100 | 1.2948 | 0.4141 |
103
- | No log | 12.5 | 200 | 1.2455 | 0.4747 |
104
- | No log | 18.75 | 300 | 1.3863 | 0.4646 |
105
- | No log | 25.0 | 400 | 1.5319 | 0.4545 |
106
 
107
 
108
  ### Framework versions
 
1
  ---
2
  license: apache-2.0
3
+ base_model: zainulhakim/240615-wav2vec2-ASR-Arabs
4
  tags:
5
  - generated_from_trainer
6
  metrics:
 
61
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/projektarbeit_dalim/huggingface/runs/ogklm4ck)
62
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/projektarbeit_dalim/huggingface/runs/ogklm4ck)
63
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/projektarbeit_dalim/huggingface/runs/ogklm4ck)
64
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/projektarbeit_dalim/dalim/runs/7r04i5y6)
65
  # 240624-wav2vec2-ASR-Arab
66
 
67
+ This model is a fine-tuned version of [zainulhakim/240615-wav2vec2-ASR-Arabs](https://huggingface.co/zainulhakim/240615-wav2vec2-ASR-Arabs) on an unknown dataset.
68
  It achieves the following results on the evaluation set:
69
+ - Loss: 1.6706
70
+ - Wer: 0.9697
71
 
72
  ## Model description
73
 
 
93
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
94
  - lr_scheduler_type: linear
95
  - lr_scheduler_warmup_steps: 500
96
+ - num_epochs: 20
97
  - mixed_precision_training: Native AMP
98
 
99
  ### Training results
100
 
101
  | Training Loss | Epoch | Step | Validation Loss | Wer |
102
  |:-------------:|:-----:|:----:|:---------------:|:------:|
103
+ | No log | 6.25 | 100 | 2.5459 | 1.0 |
104
+ | No log | 12.5 | 200 | 1.4102 | 1.0 |
105
+ | No log | 18.75 | 300 | 1.6706 | 0.9697 |
 
106
 
107
 
108
  ### Framework versions