senga-LUK-aligned-degate60-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: 0.1228
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
- train_batch_size: 8
- eval_batch_size: 8
- seed: 3407
- 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: cosine
- lr_scheduler_warmup_steps: 4000
- training_steps: 40000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.0981 | 90.9091 | 1000 | 0.0815 |
| 0.0783 | 181.8182 | 2000 | 0.0771 |
| 0.0675 | 272.7273 | 3000 | 0.0795 |
| 0.0637 | 363.6364 | 4000 | 0.0816 |
| 0.0578 | 454.5455 | 5000 | 0.0845 |
| 0.054 | 545.4545 | 6000 | 0.0882 |
| 0.0491 | 636.3636 | 7000 | 0.0898 |
| 0.0449 | 727.2727 | 8000 | 0.0959 |
| 0.0459 | 818.1818 | 9000 | 0.0961 |
| 0.0414 | 909.0909 | 10000 | 0.0986 |
| 0.0394 | 1000.0 | 11000 | 0.1011 |
| 0.0408 | 1090.9091 | 12000 | 0.1005 |
| 0.0378 | 1181.8182 | 13000 | 0.1051 |
| 0.038 | 1272.7273 | 14000 | 0.1058 |
| 0.0345 | 1363.6364 | 15000 | 0.1077 |
| 0.0355 | 1454.5455 | 16000 | 0.1095 |
| 0.0346 | 1545.4545 | 17000 | 0.1112 |
| 0.0444 | 1636.3636 | 18000 | 0.1116 |
| 0.0329 | 1727.2727 | 19000 | 0.1144 |
| 0.0344 | 1818.1818 | 20000 | 0.1143 |
| 0.0304 | 1909.0909 | 21000 | 0.1164 |
| 0.0313 | 2000.0 | 22000 | 0.1165 |
| 0.0304 | 2090.9091 | 23000 | 0.1164 |
| 0.0308 | 2181.8182 | 24000 | 0.1188 |
| 0.0299 | 2272.7273 | 25000 | 0.1208 |
| 0.0294 | 2363.6364 | 26000 | 0.1191 |
| 0.0275 | 2454.5455 | 27000 | 0.1214 |
| 0.0269 | 2545.4545 | 28000 | 0.1206 |
| 0.0283 | 2636.3636 | 29000 | 0.1207 |
| 0.0292 | 2727.2727 | 30000 | 0.1208 |
| 0.0364 | 2818.1818 | 31000 | 0.1222 |
| 0.0268 | 2909.0909 | 32000 | 0.1229 |
| 0.0261 | 3000.0 | 33000 | 0.1203 |
| 0.028 | 3090.9091 | 34000 | 0.1216 |
| 0.0257 | 3181.8182 | 35000 | 0.1218 |
| 0.0253 | 3272.7273 | 36000 | 0.1227 |
| 0.0275 | 3363.6364 | 37000 | 0.1230 |
| 0.0264 | 3454.5455 | 38000 | 0.1215 |
| 0.0274 | 3545.4545 | 39000 | 0.1223 |
| 0.0272 | 3636.3636 | 40000 | 0.1228 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.8.0+cu128
- Datasets 4.2.0
- Tokenizers 0.22.2