speecht5_finetuned_sample

This model is a fine-tuned version of microsoft/speecht5_tts on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6517

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: 1500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.581	88.8889	100	0.6059
0.4879	177.7778	200	0.5648
0.4272	266.6667	300	0.5928
0.3992	355.5556	400	0.6150
0.3642	444.4444	500	0.5903
0.349	533.3333	600	0.6436
0.3298	622.2222	700	0.6391
0.3178	711.1111	800	0.6306
0.3041	800.0	900	0.6405
0.2981	888.8889	1000	0.6490
0.2913	977.7778	1100	0.6502
0.2846	1066.6667	1200	0.6444
0.2873	1155.5556	1300	0.6636
0.2857	1244.4444	1400	0.6504
0.2839	1333.3333	1500	0.6517

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 3.1.0
• Tokenizers 0.19.1