metadata
license: apache-2.0
tags:
- generated_from_trainer
base_model: facebook/wav2vec2-base
datasets:
- transcribed_calls
metrics:
- wer
model-index:
- name: wav2vec2-base-wonders-phonemes
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: transcribed_calls
type: transcribed_calls
config: default
split: None
args: default
metrics:
- type: wer
value: 1
name: Wer
wav2vec2-base-wonders-phonemes
This model is a fine-tuned version of facebook/wav2vec2-base on the transcribed_calls dataset. It achieves the following results on the evaluation set:
- Loss: 4.1008
- Wer: 1.0
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.5174 | 0.73 | 2000 | 4.1615 | 1.0 |
| 3.3943 | 1.45 | 4000 | 4.2182 | 1.0 |
| 3.3465 | 2.18 | 6000 | 4.2119 | 1.0 |
| 3.3343 | 2.9 | 8000 | 4.2471 | 1.0 |
| 3.3546 | 3.63 | 10000 | 4.0944 | 1.0 |
| 3.7063 | 4.36 | 12000 | 4.1280 | 1.0 |
| 3.352 | 5.08 | 14000 | 4.0348 | 1.0 |
| 3.4307 | 5.81 | 16000 | 4.1120 | 1.0 |
| 3.3496 | 6.54 | 18000 | 4.0892 | 1.0 |
| 3.3681 | 7.26 | 20000 | 4.1087 | 1.0 |
| 3.3285 | 7.99 | 22000 | 4.0877 | 1.0 |
| 3.4597 | 8.71 | 24000 | 4.1041 | 1.0 |
| 3.3582 | 9.44 | 26000 | 4.0729 | 1.0 |
| 3.3898 | 10.17 | 28000 | 4.0884 | 1.0 |
| 3.4005 | 10.89 | 30000 | 4.1275 | 1.0 |
| 3.3362 | 11.62 | 32000 | 4.0973 | 1.0 |
| 3.4801 | 12.35 | 34000 | 4.0916 | 1.0 |
| 3.3902 | 13.07 | 36000 | 4.0946 | 1.0 |
| 3.3554 | 13.8 | 38000 | 4.1222 | 1.0 |
| 3.5212 | 14.52 | 40000 | 4.0763 | 1.0 |
| 3.3771 | 15.25 | 42000 | 4.1347 | 1.0 |
| 3.333 | 15.98 | 44000 | 4.1217 | 1.0 |
| 3.3605 | 16.7 | 46000 | 4.1400 | 1.0 |
| 3.3526 | 17.43 | 48000 | 4.1099 | 1.0 |
| 3.3134 | 18.16 | 50000 | 4.1327 | 1.0 |
| 3.5681 | 18.88 | 52000 | 4.1114 | 1.0 |
| 3.3497 | 19.61 | 54000 | 4.1006 | 1.0 |
| 4.0995 | 20.33 | 56000 | 4.0972 | 1.0 |
| 3.3377 | 21.06 | 58000 | 4.1035 | 1.0 |
| 3.355 | 21.79 | 60000 | 4.1022 | 1.0 |
| 3.5084 | 22.51 | 62000 | 4.1252 | 1.0 |
| 3.3558 | 23.24 | 64000 | 4.1028 | 1.0 |
| 3.3328 | 23.97 | 66000 | 4.0976 | 1.0 |
| 3.8824 | 24.69 | 68000 | 4.1290 | 1.0 |
| 3.3406 | 25.42 | 70000 | 4.1173 | 1.0 |
| 3.9972 | 26.14 | 72000 | 4.1178 | 1.0 |
| 3.4915 | 26.87 | 74000 | 4.0926 | 1.0 |
| 3.5408 | 27.6 | 76000 | 4.1069 | 1.0 |
| 3.6864 | 28.32 | 78000 | 4.1030 | 1.0 |
| 3.326 | 29.05 | 80000 | 4.1058 | 1.0 |
| 3.3171 | 29.77 | 82000 | 4.1008 | 1.0 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.3.0.dev20240314+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2