You can deploy this model and start earning money today!

Makkoen/
whisper-large-v3-cit-do015-wd0-lr1e-06-steps500-PHI

Name: whisper-large-v3-cit-do015-wd0-lr1e-06-steps500-PHI
Availability: InStock

tensorboard

safetensors

whisper

generated_from_trainer

openai/whisper-large-v3

finetune

apache-2.0

Views

No views yet

FULL6

This model is a fine-tuned version of openai/whisper-large-v3 on the 970 FULL-2024-11-22 dataset. It achieves the following results on the evaluation set:

Loss: 0.3523
Wer Ortho: 19.9273
Wer: 14.7053

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-06
train_batch_size: 4
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
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: 200
training_steps: 500
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.7795	1.8265	100	0.4587	25.2085	19.0183
0.4482	3.6530	200	0.3704	21.8730	16.0813
0.3465	5.4795	300	0.3547	20.2694	14.7053
0.2946	7.3059	400	0.3511	19.7563	14.4383
0.2691	9.1324	500	0.3523	19.9273	14.7053

Framework versions

Transformers 4.44.0
Pytorch 1.13.1+cu117
Datasets 2.21.0
Tokenizers 0.19.1

Makkoen/whisper-large-v3-cit-do015-wd0-lr1e-06-steps500-PHI

FULL6

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Makkoen/
whisper-large-v3-cit-do015-wd0-lr1e-06-steps500-PHI