Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
English
whisper
hf-asr-leaderboard
Generated from Trainer
Eval Results (legacy)
# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("Ru3ll/dsn_afrispeech-shuffle")
model = AutoModelForSpeechSeq2Seq.from_pretrained("Ru3ll/dsn_afrispeech-shuffle")Quick Links
Ru3ll/dsn_afrispeech3
This model is a fine-tuned version of ru3ll/dsn_afrispeech2/whisper-small on the Afrispeech-200 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5952
- Wer: 23.8251
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: 3.46e-06
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- training_steps: 498
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.6516 | 0.5 | 249 | 0.6065 | 23.3294 |
| 0.8152 | 1.0 | 498 | 0.5952 | 23.8251 |
Framework versions
- Transformers 4.28.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
- Downloads last month
- 2
Dataset used to train Ru3ll/dsn_afrispeech-shuffle
Evaluation results
- Wer on Afrispeech-200self-reported23.825
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Ru3ll/dsn_afrispeech-shuffle")