# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("shadow-wxh/voicecommand-german-medium")
model = AutoModelForSpeechSeq2Seq.from_pretrained("shadow-wxh/voicecommand-german-medium")Quick Links
voicecommand-german-medium
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0060
- Wer: 0.9390
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: 8
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- training_steps: 100
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4615 | 1.0476 | 33 | 0.0203 | 1.7214 |
| 0.0121 | 2.0952 | 66 | 0.0113 | 0.9390 |
| 0.0043 | 3.1429 | 99 | 0.0060 | 0.9390 |
Framework versions
- Transformers 4.43.3
- Pytorch 2.4.0+cu124
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 3
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="shadow-wxh/voicecommand-german-medium")