Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Thai
whisper
Generated from Trainer
Eval Results (legacy)
# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("fruk19/E_ASR")
model = AutoModelForSpeechSeq2Seq.from_pretrained("fruk19/E_ASR")Quick Links
South_asri
This model is a fine-tuned version of openai/whisper-small on the aicookcook dataset. It achieves the following results on the evaluation set:
- Loss: 0.0603
- Wer: 6.9375
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 99
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0953 | 1.0 | 3000 | 0.0930 | 12.2729 |
| 0.0352 | 2.0 | 6000 | 0.0640 | 7.6489 |
| 0.013 | 3.0 | 9000 | 0.0603 | 6.9375 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.0.1+cu117
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for fruk19/E_ASR
Base model
openai/whisper-smallEvaluation results
- Wer on aicookcookself-reported6.937
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="fruk19/E_ASR")