Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
whisper-tiny-til-asr – AI Model by fishjojo1 | AlphaNeural AI
You can deploy this model and start earning money today!
fishjojo1
/
whisper-tiny-til-asr
like
0
transformers
tensorboard
safetensors
whisper
automatic-speech-recognition
generated_from_trainer
openai/whisper-tiny.en
finetune
apache-2.0
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
whisper-tiny-til-asr
This model is a fine-tuned version of
openai/whisper-tiny.en
on the None dataset. It achieves the following results on the evaluation set:
Loss: 0.0001
Wer: 0.0166
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: 3.7500000000000003e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
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: 100
training_steps: 1000
mixed_precision_training: Native AMP
Training results
Training Loss
Epoch
Step
Validation Loss
Wer
0.002
1.0152
200
0.0040
0.1163
0.0001
2.0305
400
0.0003
0.0499
0.0001
3.0457
600
0.0003
0.0499
0.0
4.0609
800
0.0001
0.0166
0.0
5.0761
1000
0.0001
0.0166
Framework versions
Transformers 4.41.0
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1
