greencookie-afk/Medical-Whisper-Large-v3

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

<a href="https://ibb.co/4YRxh82"><img src="https://i.ibb.co/wwh15S7/DALL-E-2024-10-05-20-47-54-A-doctor-in-a-modern-clinical-setting-carefully-listening-to-a-patient-s.webp" alt="DALL-E-2024-10-05-20-47-54-A-doctor-in-a-modern-clinical-setting-carefully-listening-to-a-patient-s" border="0"></a>

med-whisper-large-final

This model is a fine-tuned version of openai/whisper-large-v3 on the primock_data dataset.

Model description

Fine tuned version of whisper-large-v3 through transfer learning on Doctor/Patient consultations

Intended uses & limitations

Medical transcription

Training and evaluation data

Na0s/MedicalAugmenteddata

Training procedure

Exhaustive transfer learning

Training hyperparameters

The following hyperparameters were used during training:

• —learning_rate: 1e-05
• —trainbatchsize: 6
• —evalbatchsize: 6
• —seed: 42
• —gradientaccumulationsteps: 4
• —totaltrainbatch_size: 24
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: constantwithwarmup
• —lrschedulerwarmup_steps: 50
• —training_steps: 500
• —mixedprecisiontraining: Native AMP

Performance Overview:

Performance of foundation Whispers vs Medical Whisper on the Validation set.

Table: Performance of Whisper Medical on the Test set.

Framework versions

• —Transformers 4.42.4
• —Pytorch 2.3.1+cu121
• —Datasets 2.20.0
• —Tokenizers 0.19.1