Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Swahili
whisper
hf-asr-leaderboard
Generated from Trainer
Eval Results (legacy)
# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("EYEDOL/FROM_C3_NEW2")
model = AutoModelForSpeechSeq2Seq.from_pretrained("EYEDOL/FROM_C3_NEW2")Quick Links
ASR_FROM_C3_NEW
This model is a fine-tuned version of EYEDOL/FROM_C3_NEW1 on the Common Voice 13.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2171
- Wer: 16.7644
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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0665 | 0.6918 | 2000 | 0.2171 | 16.7644 |
Framework versions
- Transformers 4.52.4
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.2
- Downloads last month
- 2
Model tree for EYEDOL/FROM_C3_NEW2
Dataset used to train EYEDOL/FROM_C3_NEW2
Evaluation results
- Wer on Common Voice 13.0self-reported16.764
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="EYEDOL/FROM_C3_NEW2")