whisper-large-v3-turbo-CAENNAIS_GB_E3 – AI Model by Rziane | AlphaNeural AI

You can deploy this model and start earning money today!

Rziane/
whisper-large-v3-turbo-CAENNAIS_GB_E3

like

0

automatic-speech-recognition

generated_from_trainer

openai/whisper-large-v3-turbo

endpoints_compatible

Views

No views yet

whisper-large-v3-turbo-CAENNAIS_GB_E3

This model is a fine-tuned version of openai/whisper-large-v3-turbo on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.3715
Wer: 15.5018

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: 32
eval_batch_size: 16
seed: 42
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: 500
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	1.0	24	1.4852	54.6506
No log	2.0	48	0.6789	37.1519
No log	3.0	72	0.3924	16.3952
No log	4.0	96	0.3316	14.6085
No log	5.0	120	0.3412	13.7677
No log	6.0	144	0.3380	13.9254
No log	7.0	168	0.3667	14.6611
No log	8.0	192	0.3592	14.2407
No log	9.0	216	0.3561	14.3458
No log	10.0	240	0.3715	15.5018

Framework versions

Transformers 4.46.2
Pytorch 2.4.1+cu121
Datasets 3.1.0
Tokenizers 0.20.0