speecht5_tts_v1_100

This model is a fine-tuned version of microsoft/speecht5_tts on the saskia_may23_39768 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4580

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: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 90

Training results

Training Loss	Epoch	Step	Validation Loss
0.5267	2.7939	1000	0.4852
0.4925	5.5870	2000	0.4635
0.488	8.3802	3000	0.4534
0.4771	11.1733	4000	0.4528
0.4703	13.9672	5000	0.4473
0.4566	16.7603	6000	0.4435
0.4548	19.5535	7000	0.4532
0.4499	22.3466	8000	0.4473
0.456	25.1398	9000	0.4446
0.4531	27.9336	10000	0.4457
0.4479	30.7268	11000	0.4495
0.4461	33.5199	12000	0.4430
0.4384	36.3131	13000	0.4410
0.4346	39.1062	14000	0.4475
0.4399	41.9001	15000	0.4461
0.4342	44.6932	16000	0.4455
0.433	47.4864	17000	0.4486
0.4344	50.2795	18000	0.4568
0.433	53.0727	19000	0.4490
0.4318	55.8665	20000	0.4554
0.4292	58.6597	21000	0.4535
0.4289	61.4528	22000	0.4510
0.4284	64.2460	23000	0.4534
0.43	67.0391	24000	0.4489
0.4277	69.8330	25000	0.4541
0.429	72.6261	26000	0.4548
0.423	75.4193	27000	0.4612
0.4265	78.2124	28000	0.4516
0.4344	81.0056	29000	0.4584
0.4303	83.7994	30000	0.4610
0.4279	86.5926	31000	0.4562
0.428	89.3857	32000	0.4580

Framework versions

• Transformers 4.56.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.2