Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
Whisper-Retrained-Swahili – AI Model by KJnr | AlphaNeural AI
You can deploy this model and start earning money today!
KJnr
/
Whisper-Retrained-Swahili
like
0
transformers
tensorboard
safetensors
whisper
automatic-speech-recognition
generated_from_trainer
sw
mozilla-foundation/common_voice_11_0
openai/whisper-small
finetune
apache-2.0
model-index
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
Whisper Small - Kenn
This model is a fine-tuned version of
openai/whisper-small
on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
Loss: 0.4699
Wer: 30.2561
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 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
training_steps: 3000
mixed_precision_training: Native AMP
Training results
Training Loss
Epoch
Step
Validation Loss
Wer
0.4726
0.4342
1000
0.5844
37.1336
0.3537
0.8684
2000
0.4871
31.6459
0.2369
1.3026
3000
0.4699
30.2561
Framework versions
Transformers 4.49.0
Pytorch 2.5.1+cu121
Datasets 3.3.2
Tokenizers 0.21.0
