CoolFace
Modelpublic

sil-ai/senga-LUK-20k-l1-speecht5

sourceHugging Facemitupdated 3mo agoView on Hugging Face
0likes9downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

senga-LUK-20k-l1-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:

  • —Loss: 0.1543

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
  • —trainbatchsize: 8
  • —evalbatchsize: 8
  • —seed: 3407
  • —gradientaccumulationsteps: 4
  • —totaltrainbatch_size: 32
  • —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • —lrschedulertype: cosine
  • —lrschedulerwarmup_steps: 2000
  • —training_steps: 20000
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation Loss
0.145790.909110000.1290
0.1273181.818220000.1296
0.1134272.727330000.1334
0.1069363.636440000.1320
0.1018454.545550000.1401
0.0972545.454560000.1381
0.0926636.363670000.1432
0.0891727.272780000.1485
0.0884818.181890000.1464
0.0837909.0909100000.1501
0.08171000.0110000.1501
0.08371090.9091120000.1509
0.07891181.8182130000.1548
0.081272.7273140000.1534
0.07671363.6364150000.1551
0.07851454.5455160000.1559
0.07751545.4545170000.1542
0.08381636.3636180000.1572
0.07771727.2727190000.1557
0.08021818.1818200000.1543

Framework versions

  • —Transformers 4.57.1
  • —Pytorch 2.8.0+cu128
  • —Datasets 4.2.0
  • —Tokenizers 0.22.2