Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
whisper-base-mixedv1.6 – AI Model by carlot | AlphaNeural AI
You can deploy this model and start earning money today!
carlot
/
whisper-base-mixedv1.6
like
0
transformers
tensorboard
safetensors
whisper
automatic-speech-recognition
generated_from_trainer
audiofolder
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
whisper-base-mixedv1.6
This model was trained from scratch on the audiofolder dataset. It achieves the following results on the evaluation set:
Loss: 0.2977
Cer: 8.4783
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: 2.5e-05
train_batch_size: 2
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 4000
mixed_precision_training: Native AMP
Training results
Training Loss
Epoch
Step
Validation Loss
Cer
0.0254
8.85
1000
0.2705
9.6739
0.0001
17.7
2000
0.2947
10.2174
0.0006
26.55
3000
0.2948
8.8587
0.0
35.4
4000
0.2977
8.4783
Framework versions
Transformers 4.36.0.dev0
Pytorch 2.1.1
Datasets 2.15.0
Tokenizers 0.15.0
