whisper-tinyfinacialKIFin

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

• Loss: 0.4763
• Wer: 51.6854

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: 1.35e-05
• train_batch_size: 8
• 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: 100
• training_steps: 600

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	0.4	50	0.8037	64.0449
No log	0.8	100	0.5715	50.0
No log	1.2	150	0.5269	45.5056
No log	1.6	200	0.5250	64.6067
No log	2.0	250	0.4775	45.5056
No log	2.4	300	0.4670	46.6292
No log	2.8	350	0.4500	41.0112
No log	3.2	400	0.4640	45.5056
No log	3.6	450	0.4673	44.9438
0.2958	4.0	500	0.4807	50.5618
0.2958	4.4	550	0.4791	50.5618
0.2958	4.8	600	0.4763	51.6854

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1