File size: 1,962 Bytes
7f1e5a1 e29bf6c 7f1e5a1 e29bf6c 7f1e5a1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 |
---
license: apache-2.0
base_model: openai/whisper-tiny.en
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisperFinetune
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# whisperFinetune
This model is a fine-tuned version of [openai/whisper-tiny.en](https://huggingface.co/openai/whisper-tiny.en) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5745
- Wer: 28.7011
## 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: 0.001
- train_batch_size: 128
- 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: 100
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:-------:|
| 2.8818 | 0.2778 | 10 | 1.9210 | 39.4679 |
| 0.7557 | 0.5556 | 20 | 0.6926 | 27.2926 |
| 0.5718 | 0.8333 | 30 | 0.5717 | 23.7559 |
| 0.421 | 1.1111 | 40 | 0.5161 | 21.5023 |
| 0.3088 | 1.3889 | 50 | 0.5103 | 21.0955 |
| 0.3415 | 1.6667 | 60 | 0.5155 | 21.6901 |
| 0.3434 | 1.9444 | 70 | 0.5176 | 30.0156 |
| 0.1371 | 2.2222 | 80 | 0.5303 | 20.6886 |
| 0.1349 | 2.5 | 90 | 0.5589 | 22.0344 |
| 0.1461 | 2.7778 | 100 | 0.5745 | 28.7011 |
### Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.1.dev0
- Tokenizers 0.19.1
|