dgo-tts-training-data-a-speecht5

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

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 4000
training_steps: 40000
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
0.6491	5.3763	1000	0.6131
0.5683	10.7527	2000	0.5488
0.543	16.1290	3000	0.5120
0.5203	21.5054	4000	0.5201
0.5149	26.8817	5000	0.5136
0.4962	32.2581	6000	0.4971
0.488	37.6344	7000	0.4969
0.4742	43.0108	8000	0.4880
0.4746	48.3871	9000	0.4941
0.4634	53.7634	10000	0.4886
0.4491	59.1398	11000	0.4912
0.4416	64.5161	12000	0.4854
0.4388	69.8925	13000	0.4894
0.4263	75.2688	14000	0.4911
0.425	80.6452	15000	0.4853
0.4179	86.0215	16000	0.4862
0.4205	91.3978	17000	0.4882
0.4087	96.7742	18000	0.4869
0.4079	102.1505	19000	0.4898
0.4142	107.5269	20000	0.4892
0.4132	112.9032	21000	0.4937
0.4126	118.2796	22000	0.4908
0.4091	123.6559	23000	0.4901
0.4078	129.0323	24000	0.4886
0.4167	134.4086	25000	0.4926
0.4052	139.7849	26000	0.4906
0.4057	145.1613	27000	0.4905
0.408	150.5376	28000	0.4919
0.4054	155.9140	29000	0.4896
0.4096	161.2903	30000	0.4920
0.4087	166.6667	31000	0.4922
0.3987	172.0430	32000	0.4911
0.4006	177.4194	33000	0.4934
0.4017	182.7957	34000	0.4936
0.4	188.1720	35000	0.4923
0.401	193.5484	36000	0.4941
0.3984	198.9247	37000	0.4946
0.401	204.3011	38000	0.4923
0.4012	209.6774	39000	0.4945
0.396	215.0538	40000	0.4943

Safetensors

Model size

0.1B params

Tensor type

F32

Base model

Finetuned

(1264)

this model