KasuleTrevor/whisper-lingala-small-test

Whisper Small Hi - Sanchit Gandhi

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

• Loss: 1.6526
• Wer: 53.5292

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-05
• 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: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0002	153.8462	1000	1.4799	52.3072
0.0001	307.6923	2000	1.5528	52.1298
0.0	461.5385	3000	1.6058	52.3033
0.0	615.3846	4000	1.6294	52.8546
0.0	769.2308	5000	1.6526	53.5292

Framework versions

• Transformers 4.44.2
• Pytorch 2.1.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1