metadata
language:
- en
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- openai/whisper-small
metrics:
- wer
model-index:
- name: Whisper Small fine tuned with comms
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: BrainHack ASR
type: openai/whisper-small
metrics:
- name: Wer
type: wer
value: 0.004561529393675362
Whisper Small fine tuned with comms
This model is a fine-tuned version of openai/whisper-small on the BrainHack ASR dataset. It achieves the following results on the evaluation set:
- Loss: 0.0818
- Wer: 0.0046
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0022 | 6.4935 | 1000 | 0.0641 | 0.0057 |
| 0.0009 | 12.9870 | 2000 | 0.0705 | 0.0050 |
| 0.0 | 19.4805 | 3000 | 0.0766 | 0.0045 |
| 0.0 | 25.9740 | 4000 | 0.0805 | 0.0046 |
| 0.0 | 32.4675 | 5000 | 0.0818 | 0.0046 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1