sgangireddy/whisper-medium-hu

openai/whisper-medium

This model is a fine-tuned version of openai/whisper-medium on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3069
• Wer: 19.8399

Model description

The Model is fine-tuned for 1000 steps/updates on CV11 Hungarian train+valiation data.

• Zero-shot - 26.9 (CV9 test data, even on CV11 the WER is closer a bit higher than this)
• Fine-tuned - 19.83 (CV11 test data)

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: 64
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0031	7.46	1000	0.3069	19.8399

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0+cu117
• Datasets 2.7.1.dev0
• Tokenizers 0.13.2