metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
datasets:
- minds14
metrics:
- wer
model-index:
- name: poly-6
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: minds14
type: minds14
config: en-US
split: train[:10]
args: en-US
metrics:
- name: Wer
type: wer
value: 1
poly-6
This model is a fine-tuned version of facebook/wav2vec2-base on the minds14 dataset. It achieves the following results on the evaluation set:
- Loss: 27.5287
- 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: 0.0003
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.01
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 15.5918 | 1.0 | 1 | 27.5287 | 1.0 |
| 15.5936 | 2.0 | 2 | 27.5287 | 1.0 |
| 15.6148 | 3.0 | 3 | 27.5287 | 1.0 |
| 15.6283 | 4.0 | 4 | 27.5287 | 1.0 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.1.2+cpu
- Datasets 2.19.1
- Tokenizers 0.19.1