steja/whisper-small-shona

Whisper small shona

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

• Loss: 1.1220
• Wer: 49.9096

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: 8
• eval_batch_size: 16
• seed: 42
• distributed_type: multi-GPU
• num_devices: 3
• gradient_accumulation_steps: 2
• total_train_batch_size: 48
• total_eval_batch_size: 48
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0064	24.24	400	0.9630	50.7233
0.001	48.48	800	1.0617	49.9397
0.0005	72.73	1200	1.1016	49.9397
0.0004	96.97	1600	1.1220	49.9096
0.0003	121.21	2000	1.1298	50.0422

Framework versions

• Transformers 4.37.1
• Pytorch 1.12.0+cu102
• Datasets 2.16.1
• Tokenizers 0.15.1