Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
English
whisper
Generated from Trainer
Eval Results (legacy)
# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("Agneev/whisper-tiny-en")
model = AutoModelForSpeechSeq2Seq.from_pretrained("Agneev/whisper-tiny-en")Quick Links
Whisper Tiny EN-US - Agneev M
This model is a fine-tuned version of openai/whisper-tiny on the Minds 14 dataset. It achieves the following results on the evaluation set:
- Loss: 0.6326
- Wer Ortho: 0.3668
- Wer: 0.3488
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: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 5
- training_steps: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.3661 | 14.29 | 50 | 0.6326 | 0.3668 | 0.3488 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
- Downloads last month
- 5
Dataset used to train Agneev/whisper-tiny-en
Evaluation results
- Wer on Minds 14self-reported0.349
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Agneev/whisper-tiny-en")