speecht5_finetuned

This model is a fine-tuned version of microsoft/speecht5_tts on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.5239

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: 2e-05
train_batch_size: 4
eval_batch_size: 4
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: 100
num_epochs: 25
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.1364	100	0.7136
0.826	2.2727	200	0.5898
0.826	3.4091	300	0.5733
0.6254	4.5455	400	0.5604
0.6254	5.6818	500	0.5542
0.603	6.8182	600	0.5490
0.603	7.9545	700	0.5450
0.5924	9.0909	800	0.5432
0.5924	10.2273	900	0.5403
0.5841	11.3636	1000	0.5378
0.5841	12.5	1100	0.5336
0.578	13.6364	1200	0.5357
0.578	14.7727	1300	0.5321
0.5724	15.9091	1400	0.5293
0.5724	17.0455	1500	0.5287
0.5704	18.1818	1600	0.5272
0.5704	19.3182	1700	0.5281
0.5653	20.4545	1800	0.5239
0.5653	21.5909	1900	0.5276
0.5623	22.7273	2000	0.5260
0.5623	23.8636	2100	0.5233
0.5628	25.0	2200	0.5239

Framework versions

Transformers 4.51.3
Pytorch 2.7.1+cu118
Datasets 3.6.0
Tokenizers 0.21.1

speecht5_finetuned

This model is a fine-tuned version of microsoft/speecht5_tts on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.5239

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: 2e-05

train_batch_size: 4

eval_batch_size: 4

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: 100

num_epochs: 25

mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.1364	100	0.7136
0.826	2.2727	200	0.5898
0.826	3.4091	300	0.5733
0.6254	4.5455	400	0.5604
0.6254	5.6818	500	0.5542
0.603	6.8182	600	0.5490
0.603	7.9545	700	0.5450
0.5924	9.0909	800	0.5432
0.5924	10.2273	900	0.5403
0.5841	11.3636	1000	0.5378
0.5841	12.5	1100	0.5336
0.578	13.6364	1200	0.5357
0.578	14.7727	1300	0.5321
0.5724	15.9091	1400	0.5293
0.5724	17.0455	1500	0.5287
0.5704	18.1818	1600	0.5272
0.5704	19.3182	1700	0.5281
0.5653	20.4545	1800	0.5239
0.5653	21.5909	1900	0.5276
0.5623	22.7273	2000	0.5260
0.5623	23.8636	2100	0.5233
0.5628	25.0	2200	0.5239

Framework versions

Transformers 4.51.3

Pytorch 2.7.1+cu118

Datasets 3.6.0

Tokenizers 0.21.1

speecht5 finetuned