|
--- |
|
language: |
|
- da |
|
license: mit |
|
base_model: microsoft/speecht5_tts |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- alexandrainst/nst-da |
|
model-index: |
|
- name: speecht5_tts-finetuned-nst-da |
|
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. --> |
|
|
|
# speecht5_tts-finetuned-nst-da |
|
|
|
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the NST Danish ASR Database dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3387 |
|
- Mae: 0.1486 |
|
- Mse: 0.0395 |
|
- Mcd: 16.3878 |
|
- Mae Rmcd: 0.1562 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 20 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Mae | Mse | Mcd | Mae Rmcd | |
|
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:-------:|:--------:| |
|
| 0.4202 | 1.0 | 2378 | 0.3851 | 0.1589 | 0.0446 | 17.5859 | 0.1674 | |
|
| 0.3897 | 2.0 | 4756 | 0.3771 | 0.1578 | 0.0440 | 17.2975 | 0.1654 | |
|
| 0.3749 | 3.0 | 7134 | 0.3636 | 0.1605 | 0.0450 | 17.3449 | 0.1670 | |
|
| 0.359 | 4.0 | 9512 | 0.3632 | 0.1580 | 0.0443 | 17.2756 | 0.1654 | |
|
| 0.3532 | 5.0 | 11890 | 0.3539 | 0.1592 | 0.0444 | 17.1294 | 0.1653 | |
|
| 0.356 | 6.0 | 14268 | 0.3455 | 0.1536 | 0.0418 | 16.9892 | 0.1617 | |
|
| 0.3437 | 7.0 | 16646 | 0.3426 | 0.1528 | 0.0416 | 16.8165 | 0.1605 | |
|
| 0.3434 | 8.0 | 19024 | 0.3438 | 0.1534 | 0.0416 | 16.8859 | 0.1611 | |
|
| 0.3391 | 9.0 | 21402 | 0.3487 | 0.1558 | 0.0429 | 16.7245 | 0.1615 | |
|
| 0.3381 | 10.0 | 23780 | 0.3414 | 0.1516 | 0.0409 | 16.6136 | 0.1589 | |
|
| 0.3455 | 11.0 | 26158 | 0.3395 | 0.1509 | 0.0406 | 16.6252 | 0.1586 | |
|
| 0.3408 | 12.0 | 28536 | 0.3402 | 0.1504 | 0.0404 | 16.7043 | 0.1587 | |
|
| 0.341 | 13.0 | 30914 | 0.3420 | 0.1527 | 0.0414 | 16.6290 | 0.1595 | |
|
| 0.3279 | 14.0 | 33292 | 0.3388 | 0.1503 | 0.0404 | 16.5525 | 0.1579 | |
|
| 0.328 | 15.0 | 35670 | 0.3389 | 0.1498 | 0.0401 | 16.4898 | 0.1574 | |
|
| 0.3264 | 16.0 | 38048 | 0.3391 | 0.1497 | 0.0400 | 16.4882 | 0.1573 | |
|
| 0.3318 | 17.0 | 40426 | 0.3383 | 0.1498 | 0.0402 | 16.5091 | 0.1575 | |
|
| 0.3299 | 18.0 | 42804 | 0.3399 | 0.1492 | 0.0399 | 16.4604 | 0.1569 | |
|
| 0.3304 | 19.0 | 45182 | 0.3399 | 0.1489 | 0.0396 | 16.4132 | 0.1565 | |
|
| 0.3204 | 20.0 | 47560 | 0.3387 | 0.1486 | 0.0395 | 16.3878 | 0.1562 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.37.2 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.17.0 |
|
- Tokenizers 0.15.2 |
|
|
