whisper-new – AI Model by randaaouini | AlphaNeural AI

You can deploy this model and start earning money today!

randaaouini/
whisper-new

like

0

automatic-speech-recognition

generated_from_trainer

openai/whisper-medium

endpoints_compatible

Views

No views yet

whisper-new

This model is a fine-tuned version of openai/whisper-medium on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0770
Wer: 0.8251
Cer: 1.5077

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: 5e-05
train_batch_size: 4
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Use OptimizerNames.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: 20
training_steps: 100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
1.152	0.3008	20	0.7840	2.2045	1.7712
0.4326	0.6015	40	0.2749	1.1522	1.2339
0.3414	0.9023	60	0.1763	2.8014	4.7745
0.1267	1.1955	80	0.1082	1.0049	1.2254
0.1284	1.4962	100	0.0770	0.8251	1.5077

Framework versions

Transformers 4.48.3
Pytorch 2.4.1+cu121
Datasets 3.6.0
Tokenizers 0.21.0