# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForCTC
processor = AutoProcessor.from_pretrained("Sjdan/m11_main")
model = AutoModelForCTC.from_pretrained("Sjdan/m11_main", device_map="auto")Quick Links
m11_main
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0801
- Wer: 1.6706
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: 0.0001
- train_batch_size: 8
- 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: 1000
- num_epochs: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 9.1429 | 0.96 | 500 | 3.1312 | 1.0 |
| 2.5654 | 1.92 | 1000 | 1.6310 | 2.8323 |
| 1.0072 | 2.88 | 1500 | 0.3978 | 2.0118 |
| 0.3043 | 3.84 | 2000 | 0.1985 | 1.3826 |
| 0.1542 | 4.8 | 2500 | 0.1338 | 1.7140 |
| 0.0791 | 5.76 | 3000 | 0.0935 | 1.5720 |
| 0.0594 | 6.72 | 3500 | 0.0801 | 1.6706 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.13.1+cu116
- Datasets 1.18.3
- Tokenizers 0.13.2
- Downloads last month
- 3
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Sjdan/m11_main")