You can deploy this model and start earning money today!

VladS159/
whisper-medium-6000-steps_15_percent_4_speakers_synthetic_data_24_02_2026

Name: whisper-medium-6000-steps_15_percent_4_speakers_synthetic_data_24_02_2026
Availability: InStock

VladS159/romanian_speech_dataset_with_15_percent_4_speakers_synthetic_data

openai/whisper-medium

apache-2.0

model-index

Views

No views yet

Whisper Medium Ro - PEFT

This model is a fine-tuned version of openai/whisper-medium on the Common Voice 17.0 15% Synthetic 4 Speakers dataset. It achieves the following results on the evaluation set:

Loss: 0.1622
Wer: 8.3578

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: 8
eval_batch_size: 8
seed: 42
optimizer: Use adamw_torch_fused 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: 600
training_steps: 6000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2446	0.2267	1000	0.2123	10.6319
0.2134	0.4533	2000	0.1823	9.4719
0.1992	0.6800	3000	0.1724	8.8187
0.1934	0.9066	4000	0.1671	8.4921
0.2008	1.1333	5000	0.1629	8.3181
0.1698	1.3599	6000	0.1622	8.3578

Framework versions

PEFT 0.18.1.dev0
Transformers 4.57.1
Pytorch 2.9.1+rocm6.4
Datasets 3.6.0
Tokenizers 0.22.1

VladS159/whisper-medium-6000-steps_15_percent_4_speakers_synthetic_data_24_02_2026

Whisper Medium Ro - PEFT

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

VladS159/
whisper-medium-6000-steps_15_percent_4_speakers_synthetic_data_24_02_2026