speecht5_finetuned_emirhan_tr

This model is a fine-tuned version of aarishshahmohsin/speecht5_finetuned_emirhan_tr on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4764

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• 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: 100
• training_steps: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5298	0.3772	100	0.5179
0.5389	0.7544	200	0.5179
0.5313	1.1315	300	0.5084
0.5295	1.5087	400	0.5084
0.5198	1.8859	500	0.5086
0.5225	2.2631	600	0.5081
0.5166	2.6403	700	0.5023
0.5128	3.0174	800	0.5004
0.5071	3.3946	900	0.4976
0.5045	3.7718	1000	0.4923
0.4969	4.1490	1100	0.4944
0.5023	4.5262	1200	0.4854
0.5008	4.9033	1300	0.4893
0.4973	5.2805	1400	0.4886
0.4916	5.6577	1500	0.4841
0.4915	6.0349	1600	0.4801
0.4799	6.4121	1700	0.4799
0.4832	6.7893	1800	0.4766
0.4886	7.1664	1900	0.4776
0.474	7.5436	2000	0.4764

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0