Commit
•
7b1b80d
1
Parent(s):
c319d63
Model save
Browse files
README.md
CHANGED
|
@@ -2,8 +2,6 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
| 4 |
tags:
|
| 5 |
-
- automatic-speech-recognition
|
| 6 |
-
- DewiBrynJones/banc-trawsgrifiadau-bangor-clean
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- wer
|
|
@@ -17,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
# wav2vec2-xlsr-53-ft-btb-cy
|
| 19 |
|
| 20 |
-
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
-
- Loss: 0.
|
| 23 |
-
- Wer: 0.
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
@@ -40,29 +38,25 @@ More information needed
|
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
| 42 |
- learning_rate: 0.0003
|
| 43 |
-
- train_batch_size:
|
| 44 |
-
- eval_batch_size:
|
| 45 |
- seed: 42
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
-
- lr_scheduler_warmup_steps:
|
| 49 |
-
- training_steps:
|
| 50 |
- mixed_precision_training: Native AMP
|
| 51 |
|
| 52 |
### Training results
|
| 53 |
|
| 54 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 55 |
|:-------------:|:------:|:----:|:---------------:|:------:|
|
| 56 |
-
| No log | 0.
|
| 57 |
-
| No log | 0.
|
| 58 |
-
| 2.
|
| 59 |
-
| 2.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.8304 | 1.5945 | 1400 | 0.5639 | 0.4210 |
|
| 63 |
-
| 0.71 | 1.8223 | 1600 | 0.5502 | 0.4186 |
|
| 64 |
-
| 0.71 | 2.0501 | 1800 | 0.5441 | 0.4073 |
|
| 65 |
-
| 0.6605 | 2.2779 | 2000 | 0.5418 | 0.3991 |
|
| 66 |
|
| 67 |
|
| 68 |
### Framework versions
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- wer
|
|
|
|
| 15 |
|
| 16 |
# wav2vec2-xlsr-53-ft-btb-cy
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.7008
|
| 21 |
+
- Wer: 0.4816
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
- learning_rate: 0.0003
|
| 41 |
+
- train_batch_size: 64
|
| 42 |
+
- eval_batch_size: 64
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
+
- lr_scheduler_warmup_steps: 120
|
| 47 |
+
- training_steps: 1200
|
| 48 |
- mixed_precision_training: Native AMP
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 53 |
|:-------------:|:------:|:----:|:---------------:|:------:|
|
| 54 |
+
| No log | 0.4556 | 200 | 2.5809 | 0.9998 |
|
| 55 |
+
| No log | 0.9112 | 400 | 0.8879 | 0.6325 |
|
| 56 |
+
| 2.4327 | 1.3667 | 600 | 0.6609 | 0.4897 |
|
| 57 |
+
| 2.4327 | 1.8223 | 800 | 0.6689 | 0.5018 |
|
| 58 |
+
| 0.8178 | 2.2779 | 1000 | 0.6797 | 0.4852 |
|
| 59 |
+
| 0.8178 | 2.7335 | 1200 | 0.7008 | 0.4816 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|