README.md

metadata

base_model: openai/whisper-medium
datasets:
  - facebook/voxpopuli
language:
  - it
library_name: peft
license: apache-2.0
metrics:
  - wer
tags:
  - generated_from_trainer
model-index:
  - name: Whisper Medium
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: facebook/voxpopuli
          type: facebook/voxpopuli
          config: it
          split: None
          args: it
        metrics:
          - type: wer
            value: 7.118604378878351
            name: Wer

Whisper Medium

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

• Loss: 0.1554
• Wer: 7.1186

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: 0.0001
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 200
• training_steps: 1200
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.169	0.0762	400	0.1676	7.7743
0.1679	0.1523	800	0.1833	7.2357
0.1584	0.2285	1200	0.1554	7.1186

Framework versions

• PEFT 0.12.0
• Transformers 4.43.1
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1