naqi72/speecht5_finetuned_tech_words_syed_naqi

SyedNaqi_TechSpeechT5_TTS

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

• Loss: 0.4548

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: 5e-05
• train_batch_size: 12
• eval_batch_size: 10
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 24
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5565	1.3423	500	0.4948
0.5157	2.6846	1000	0.4702
0.4906	4.0268	1500	0.4601
0.4767	5.3691	2000	0.4567
0.4718	6.7114	2500	0.4531
0.4609	8.0537	3000	0.4533
0.4545	9.3960	3500	0.4543
0.4474	10.7383	4000	0.4544
0.4505	12.0805	4500	0.4546
0.4376	13.4228	5000	0.4548

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Tokenizers 0.19.1