Japanese_Fine_Tuned_Whisper_Model

This model is a fine-tuned version of openai/whisper-tiny on the Common Voice dataset. It achieves the following results on the evaluation set:

• Loss: 0.549100
• Wer: 225.233037

Model description

The tiny Whisper model is fine-tuned on Japanese speech samples from the Common Voice dataset, based on which users can perform Automatic Speech Recognition in real time in Japanese.

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Step	Validation Loss	Wer
0.8097	200	0.801917	601.560806
0.7200	400	0.783436	327.335790
0.6810	600	0.759281	254.064600
0.7351	800	0.747759	241.426404
0.5491	1000	0.747127	225.233037

Framework versions

• Transformers 4.27.0.dev0
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2