Commit
•
08d5a50
1
Parent(s):
0ee778a
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:
|
| 23 |
-
- Wer: 0.
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
@@ -40,49 +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: 8
|
| 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: 500
|
| 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 |
-
| 3.
|
| 59 |
-
| 3.
|
| 60 |
-
|
|
| 61 |
-
|
|
| 62 |
-
| 1.0665 | 0.7973 | 1400 | 0.6942 | 0.5389 |
|
| 63 |
-
| 0.8811 | 0.9112 | 1600 | 0.6152 | 0.4862 |
|
| 64 |
-
| 0.8811 | 1.0251 | 1800 | 0.5941 | 0.4535 |
|
| 65 |
-
| 0.7654 | 1.1390 | 2000 | 0.5642 | 0.4329 |
|
| 66 |
-
| 0.7654 | 1.2528 | 2200 | 0.5727 | 0.4313 |
|
| 67 |
-
| 0.7654 | 1.3667 | 2400 | 0.5467 | 0.4317 |
|
| 68 |
-
| 0.6896 | 1.4806 | 2600 | 0.5658 | 0.4398 |
|
| 69 |
-
| 0.6896 | 1.5945 | 2800 | 0.6008 | 0.4445 |
|
| 70 |
-
| 0.7558 | 1.7084 | 3000 | 0.7090 | 0.5021 |
|
| 71 |
-
| 0.7558 | 1.8223 | 3200 | 0.8050 | 0.5291 |
|
| 72 |
-
| 0.7558 | 1.9362 | 3400 | 0.8248 | 0.5275 |
|
| 73 |
-
| 0.9823 | 2.0501 | 3600 | 0.7945 | 0.5073 |
|
| 74 |
-
| 0.9823 | 2.1640 | 3800 | 0.7690 | 0.4982 |
|
| 75 |
-
| 0.9181 | 2.2779 | 4000 | 0.8669 | 0.6930 |
|
| 76 |
-
| 0.9181 | 2.3918 | 4200 | 0.8887 | 0.6311 |
|
| 77 |
-
| 0.9181 | 2.5057 | 4400 | 1.1509 | 0.8436 |
|
| 78 |
-
| 1.0805 | 2.6196 | 4600 | 1.2167 | 0.9459 |
|
| 79 |
-
| 1.0805 | 2.7335 | 4800 | 1.2323 | 0.9307 |
|
| 80 |
-
| 1.3977 | 2.8474 | 5000 | 1.6597 | 0.9528 |
|
| 81 |
-
| 1.3977 | 2.9613 | 5200 | 2.0224 | 0.9162 |
|
| 82 |
-
| 1.3977 | 3.0752 | 5400 | 2.1054 | 0.9104 |
|
| 83 |
-
| 2.1028 | 3.1891 | 5600 | 2.2486 | 0.9281 |
|
| 84 |
-
| 2.1028 | 3.3030 | 5800 | 2.2447 | 0.9216 |
|
| 85 |
-
| 2.2911 | 3.4169 | 6000 | 2.2310 | 0.9240 |
|
| 86 |
|
| 87 |
|
| 88 |
### 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: 0.5375
|
| 21 |
+
- Wer: 0.4141
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
- learning_rate: 0.0003
|
| 41 |
+
- train_batch_size: 32
|
| 42 |
- eval_batch_size: 8
|
| 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: 500
|
| 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 | 0.2278 | 200 | 2.9538 | 1.0 |
|
| 55 |
+
| No log | 0.4556 | 400 | 1.4731 | 0.8217 |
|
| 56 |
+
| 3.3429 | 0.6834 | 600 | 0.8524 | 0.6398 |
|
| 57 |
+
| 3.3429 | 0.9112 | 800 | 0.7107 | 0.5182 |
|
| 58 |
+
| 0.8743 | 1.1390 | 1000 | 0.5668 | 0.4325 |
|
| 59 |
+
| 0.8743 | 1.3667 | 1200 | 0.5375 | 0.4141 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|