Commit
•
66985d8
1
Parent(s):
a4e31fa
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-with-ccv
|
| 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-ccv-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: inf
|
| 23 |
-
- Wer: 0.
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
@@ -40,74 +38,54 @@ 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
|
| 55 |
-
|
| 56 |
-
| No log | 0.
|
| 57 |
-
| No log | 0.
|
| 58 |
-
| 3.
|
| 59 |
-
| 3.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
|
|
| 76 |
-
|
|
| 77 |
-
|
|
| 78 |
-
|
|
| 79 |
-
|
|
| 80 |
-
|
|
| 81 |
-
|
|
| 82 |
-
|
|
| 83 |
-
|
|
| 84 |
-
|
|
| 85 |
-
|
|
| 86 |
-
| 2.8212 | 4.7876 | 6200 | inf | 0.9959 |
|
| 87 |
-
| 2.8212 | 4.9421 | 6400 | inf | 0.9917 |
|
| 88 |
-
| 2.7652 | 5.0965 | 6600 | inf | 0.9897 |
|
| 89 |
-
| 2.7652 | 5.2510 | 6800 | inf | 0.9902 |
|
| 90 |
-
| 2.7358 | 5.4054 | 7000 | inf | 0.9889 |
|
| 91 |
-
| 2.7358 | 5.5598 | 7200 | inf | 0.9905 |
|
| 92 |
-
| 2.7358 | 5.7143 | 7400 | inf | 0.9887 |
|
| 93 |
-
| 2.7122 | 5.8687 | 7600 | inf | 0.9878 |
|
| 94 |
-
| 2.7122 | 6.0232 | 7800 | inf | 0.9847 |
|
| 95 |
-
| 2.7345 | 6.1776 | 8000 | inf | 0.9842 |
|
| 96 |
-
| 2.7345 | 6.3320 | 8200 | inf | 0.9882 |
|
| 97 |
-
| 2.7345 | 6.4865 | 8400 | inf | 0.9872 |
|
| 98 |
-
| 3.035 | 6.6409 | 8600 | inf | 0.9921 |
|
| 99 |
-
| 3.035 | 6.7954 | 8800 | inf | 0.9906 |
|
| 100 |
-
| 3.688 | 6.9498 | 9000 | inf | 0.9916 |
|
| 101 |
-
| 3.688 | 7.1042 | 9200 | inf | 0.9906 |
|
| 102 |
-
| 3.688 | 7.2587 | 9400 | inf | 0.9908 |
|
| 103 |
-
| 3.7017 | 7.4131 | 9600 | inf | 0.9912 |
|
| 104 |
-
| 3.7017 | 7.5676 | 9800 | inf | 0.9913 |
|
| 105 |
-
| 3.7327 | 7.7220 | 10000 | inf | 0.9913 |
|
| 106 |
|
| 107 |
|
| 108 |
### Framework versions
|
| 109 |
|
| 110 |
-
- Transformers 4.
|
| 111 |
-
- Pytorch 2.
|
| 112 |
-
- Datasets 2.
|
| 113 |
- Tokenizers 0.19.1
|
|
|
|
| 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-ccv-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: inf
|
| 21 |
+
- Wer: 0.3264
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
- learning_rate: 0.0003
|
| 41 |
+
- train_batch_size: 16
|
| 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: 600
|
| 47 |
+
- training_steps: 6000
|
| 48 |
- mixed_precision_training: Native AMP
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 53 |
+
|:-------------:|:------:|:----:|:---------------:|:------:|
|
| 54 |
+
| No log | 0.0772 | 200 | inf | 1.0 |
|
| 55 |
+
| No log | 0.1544 | 400 | inf | 0.8963 |
|
| 56 |
+
| 3.9177 | 0.2317 | 600 | inf | 0.7595 |
|
| 57 |
+
| 3.9177 | 0.3089 | 800 | inf | 0.7512 |
|
| 58 |
+
| 0.9791 | 0.3861 | 1000 | inf | 0.5984 |
|
| 59 |
+
| 0.9791 | 0.4633 | 1200 | inf | 0.5868 |
|
| 60 |
+
| 0.9791 | 0.5405 | 1400 | inf | 0.5255 |
|
| 61 |
+
| 0.805 | 0.6178 | 1600 | inf | 0.5282 |
|
| 62 |
+
| 0.805 | 0.6950 | 1800 | inf | 0.4769 |
|
| 63 |
+
| 0.7184 | 0.7722 | 2000 | inf | 0.4743 |
|
| 64 |
+
| 0.7184 | 0.8494 | 2200 | inf | 0.4680 |
|
| 65 |
+
| 0.7184 | 0.9266 | 2400 | inf | 0.4570 |
|
| 66 |
+
| 0.6704 | 1.0039 | 2600 | inf | 0.4253 |
|
| 67 |
+
| 0.6704 | 1.0811 | 2800 | inf | 0.4164 |
|
| 68 |
+
| 0.5664 | 1.1583 | 3000 | inf | 0.4159 |
|
| 69 |
+
| 0.5664 | 1.2355 | 3200 | inf | 0.3995 |
|
| 70 |
+
| 0.5664 | 1.3127 | 3400 | inf | 0.3941 |
|
| 71 |
+
| 0.5359 | 1.3900 | 3600 | inf | 0.3819 |
|
| 72 |
+
| 0.5359 | 1.4672 | 3800 | inf | 0.3811 |
|
| 73 |
+
| 0.5172 | 1.5444 | 4000 | inf | 0.3691 |
|
| 74 |
+
| 0.5172 | 1.6216 | 4200 | inf | 0.3609 |
|
| 75 |
+
| 0.5172 | 1.6988 | 4400 | inf | 0.3600 |
|
| 76 |
+
| 0.4817 | 1.7761 | 4600 | inf | 0.3509 |
|
| 77 |
+
| 0.4817 | 1.8533 | 4800 | inf | 0.3530 |
|
| 78 |
+
| 0.4818 | 1.9305 | 5000 | inf | 0.3434 |
|
| 79 |
+
| 0.4818 | 2.0077 | 5200 | inf | 0.3363 |
|
| 80 |
+
| 0.4818 | 2.0849 | 5400 | inf | 0.3372 |
|
| 81 |
+
| 0.4196 | 2.1622 | 5600 | inf | 0.3320 |
|
| 82 |
+
| 0.4196 | 2.2394 | 5800 | inf | 0.3293 |
|
| 83 |
+
| 0.3743 | 2.3166 | 6000 | inf | 0.3264 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 84 |
|
| 85 |
|
| 86 |
### Framework versions
|
| 87 |
|
| 88 |
+
- Transformers 4.44.0
|
| 89 |
+
- Pytorch 2.4.0+cu121
|
| 90 |
+
- Datasets 2.21.0
|
| 91 |
- Tokenizers 0.19.1
|