|
--- |
|
license: apache-2.0 |
|
base_model: openai/whisper-tiny |
|
tags: |
|
- ASR |
|
- Papiamentu |
|
- Whisper |
|
- Speech Recognition |
|
- generated_from_trainer |
|
datasets: |
|
- sonnygeorge/papi_asr_corpus |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: Whisper Tiny Papiamentu |
|
results: |
|
- task: |
|
name: Automatic Speech Recognition |
|
type: automatic-speech-recognition |
|
dataset: |
|
name: Papi ASR |
|
type: sonnygeorge/papi_asr_corpus |
|
metrics: |
|
- name: Wer |
|
type: wer |
|
value: 85.22752415369865 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# Whisper Tiny Papiamentu |
|
|
|
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Papi ASR dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.0809 |
|
- Wer: 85.2275 |
|
- Cer: 34.1432 |
|
|
|
## 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: 100 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-------:| |
|
| No log | 0.01 | 5 | 4.6530 | 101.2317 | 53.1862 | |
|
| No log | 0.01 | 10 | 4.6414 | 102.1683 | 53.8408 | |
|
| No log | 0.02 | 15 | 4.6073 | 102.0577 | 53.5251 | |
|
| No log | 0.03 | 20 | 4.5651 | 101.3275 | 52.8798 | |
|
| 4.5851 | 0.04 | 25 | 4.3546 | 100.9956 | 52.4930 | |
|
| 4.5851 | 0.04 | 30 | 4.2726 | 98.9158 | 50.6731 | |
|
| 4.5851 | 0.05 | 35 | 4.1573 | 98.3922 | 50.3358 | |
|
| 4.5851 | 0.06 | 40 | 3.9879 | 96.8360 | 48.6228 | |
|
| 4.5851 | 0.07 | 45 | 3.8552 | 96.4747 | 47.4498 | |
|
| 4.1061 | 0.07 | 50 | 3.6580 | 94.7710 | 47.1001 | |
|
| 4.1061 | 0.08 | 55 | 3.4486 | 93.1780 | 45.8250 | |
|
| 4.1061 | 0.09 | 60 | 3.2790 | 94.1441 | 46.0849 | |
|
| 4.1061 | 0.1 | 65 | 3.1437 | 93.4730 | 45.1332 | |
|
| 4.1061 | 0.1 | 70 | 3.0021 | 91.7029 | 42.6202 | |
|
| 3.2929 | 0.11 | 75 | 2.8536 | 92.1086 | 41.3095 | |
|
| 3.2929 | 0.12 | 80 | 2.7078 | 90.1468 | 38.6819 | |
|
| 3.2929 | 0.13 | 85 | 2.5575 | 88.8045 | 37.2195 | |
|
| 3.2929 | 0.13 | 90 | 2.4029 | 88.6865 | 36.6887 | |
|
| 3.2929 | 0.14 | 95 | 2.2431 | 89.0257 | 35.9986 | |
|
| 2.4377 | 0.15 | 100 | 2.0809 | 85.2275 | 34.1432 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.37.0.dev0 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|