arun100/whisper-small-derived-bn-1

Whisper Base Hindi

This model is a fine-tuned version of bangla-speech-processing/BanglaASR on the mozilla-foundation/common_voice_16_0 bn dataset. It achieves the following results on the evaluation set:

• Loss: 0.1281
• Wer: 3.7265

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-07
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1588	0.5	50	0.1495	3.7650
0.1267	1.0	100	0.1281	3.7265

Framework versions

• Transformers 4.37.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.16.2.dev0
• Tokenizers 0.15.0