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: 3.9444
- 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
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 8
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.6803 | 3.17 | 200 | 4.2914 | 1.0 |
| 3.5239 | 4.76 | 300 | 4.0735 | 1.0 |
| 3.449 | 6.35 | 400 | 3.9754 | 1.0 |
| 3.9913 | 7.94 | 500 | 3.9428 | 1.0 |
| 3.3992 | 9.52 | 600 | 3.9585 | 1.0 |
| 3.3712 | 11.11 | 700 | 3.9996 | 1.0 |
| 3.3716 | 12.7 | 800 | 3.9949 | 1.0 |
| 3.9369 | 14.29 | 900 | 3.9352 | 1.0 |
| 3.3867 | 15.87 | 1000 | 3.9327 | 1.0 |
| 3.3602 | 17.46 | 1100 | 3.9940 | 1.0 |
| 3.4101 | 19.05 | 1200 | 3.9470 | 1.0 |
| 3.3484 | 20.63 | 1300 | 3.9482 | 1.0 |
| 4.4074 | 22.22 | 1400 | 3.9364 | 1.0 |
| 3.9909 | 23.81 | 1500 | 3.9681 | 1.0 |
| 3.3762 | 25.4 | 1600 | 3.9738 | 1.0 |
| 3.357 | 26.98 | 1700 | 3.9504 | 1.0 |
| 3.3729 | 28.57 | 1800 | 3.9444 | 1.0 |
Framework versions
- Transformers 4.39.1
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2