whisper-for-maltese

This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.3026
• Wer: 120.5033

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: 8
• 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: 4000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.2456	3.57	500	2.5059	99.2583
1.346	7.14	1000	2.4891	421.1391
0.7287	10.71	1500	2.7307	121.1921
0.3432	14.29	2000	2.9824	131.7086
0.2834	17.86	2500	3.1282	122.0397
0.2239	21.43	3000	3.2050	113.4040
0.1978	25.0	3500	3.2739	118.5695
0.1476	28.57	4000	3.3026	120.5033

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.14.1
• Tokenizers 0.13.3