|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
base_model: microsoft/speecht5_tts |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: CollectedDataModel |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- 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. --> |
|
|
|
|
|
# CollectedDataModel |
|
|
|
|
|
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.4390 |
|
|
|
|
|
## 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: 4 |
|
|
- eval_batch_size: 2 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 8 |
|
|
- total_train_batch_size: 32 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_steps: 100 |
|
|
- training_steps: 4000 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-------:|:----:|:---------------:| |
|
|
| 0.5567 | 0.9913 | 100 | 0.4913 | |
|
|
| 0.5127 | 1.9827 | 200 | 0.4692 | |
|
|
| 0.4915 | 2.9740 | 300 | 0.4562 | |
|
|
| 0.4862 | 3.9653 | 400 | 0.4524 | |
|
|
| 0.4745 | 4.9566 | 500 | 0.4483 | |
|
|
| 0.4735 | 5.9480 | 600 | 0.4458 | |
|
|
| 0.4681 | 6.9393 | 700 | 0.4397 | |
|
|
| 0.4656 | 7.9306 | 800 | 0.4408 | |
|
|
| 0.4576 | 8.9219 | 900 | 0.4336 | |
|
|
| 0.4571 | 9.9133 | 1000 | 0.4343 | |
|
|
| 0.451 | 10.9046 | 1100 | 0.4339 | |
|
|
| 0.4517 | 11.8959 | 1200 | 0.4316 | |
|
|
| 0.4432 | 12.8872 | 1300 | 0.4315 | |
|
|
| 0.4448 | 13.8786 | 1400 | 0.4357 | |
|
|
| 0.4455 | 14.8699 | 1500 | 0.4296 | |
|
|
| 0.4387 | 15.8612 | 1600 | 0.4331 | |
|
|
| 0.4334 | 16.8525 | 1700 | 0.4359 | |
|
|
| 0.4373 | 17.8439 | 1800 | 0.4290 | |
|
|
| 0.4304 | 18.8352 | 1900 | 0.4318 | |
|
|
| 0.4279 | 19.8265 | 2000 | 0.4305 | |
|
|
| 0.4294 | 20.8178 | 2100 | 0.4327 | |
|
|
| 0.4269 | 21.8092 | 2200 | 0.4327 | |
|
|
| 0.4248 | 22.8005 | 2300 | 0.4309 | |
|
|
| 0.4255 | 23.7918 | 2400 | 0.4275 | |
|
|
| 0.43 | 24.7831 | 2500 | 0.4315 | |
|
|
| 0.4214 | 25.7745 | 2600 | 0.4345 | |
|
|
| 0.4166 | 26.7658 | 2700 | 0.4362 | |
|
|
| 0.4173 | 27.7571 | 2800 | 0.4343 | |
|
|
| 0.4172 | 28.7485 | 2900 | 0.4325 | |
|
|
| 0.4142 | 29.7398 | 3000 | 0.4329 | |
|
|
| 0.4134 | 30.7311 | 3100 | 0.4327 | |
|
|
| 0.4121 | 31.7224 | 3200 | 0.4388 | |
|
|
| 0.4085 | 32.7138 | 3300 | 0.4352 | |
|
|
| 0.4095 | 33.7051 | 3400 | 0.4388 | |
|
|
| 0.4112 | 34.6964 | 3500 | 0.4372 | |
|
|
| 0.4106 | 35.6877 | 3600 | 0.4388 | |
|
|
| 0.4054 | 36.6791 | 3700 | 0.4392 | |
|
|
| 0.4075 | 37.6704 | 3800 | 0.4395 | |
|
|
| 0.4086 | 38.6617 | 3900 | 0.4393 | |
|
|
| 0.4125 | 39.6530 | 4000 | 0.4390 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.44.2 |
|
|
- Pytorch 2.4.1+cu121 |
|
|
- Datasets 3.0.1 |
|
|
- Tokenizers 0.19.1 |
|
|
|
