Commit
•
06df48c
1
Parent(s):
cf3ec94
update model card README.md
Browse files
README.md
CHANGED
|
@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the VOXPOPULI dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -41,22 +41,57 @@ The following hyperparameters were used during training:
|
|
| 41 |
- train_batch_size: 4
|
| 42 |
- eval_batch_size: 8
|
| 43 |
- seed: 42
|
| 44 |
-
- gradient_accumulation_steps:
|
| 45 |
-
- total_train_batch_size:
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_steps: 100
|
| 49 |
-
- num_epochs:
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
-
| Training Loss | Epoch | Step
|
| 54 |
-
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the VOXPOPULI dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.4600
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 41 |
- train_batch_size: 4
|
| 42 |
- eval_batch_size: 8
|
| 43 |
- seed: 42
|
| 44 |
+
- gradient_accumulation_steps: 4
|
| 45 |
+
- total_train_batch_size: 16
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_steps: 100
|
| 49 |
+
- num_epochs: 40
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 54 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 55 |
+
| 0.5641 | 1.0 | 712 | 0.5090 |
|
| 56 |
+
| 0.5394 | 2.0 | 1424 | 0.4915 |
|
| 57 |
+
| 0.5277 | 3.0 | 2136 | 0.4819 |
|
| 58 |
+
| 0.5136 | 4.0 | 2848 | 0.4798 |
|
| 59 |
+
| 0.5109 | 5.0 | 3560 | 0.4733 |
|
| 60 |
+
| 0.5078 | 6.0 | 4272 | 0.4731 |
|
| 61 |
+
| 0.5033 | 7.0 | 4984 | 0.4692 |
|
| 62 |
+
| 0.5021 | 8.0 | 5696 | 0.4691 |
|
| 63 |
+
| 0.4984 | 9.0 | 6408 | 0.4670 |
|
| 64 |
+
| 0.488 | 10.0 | 7120 | 0.4641 |
|
| 65 |
+
| 0.491 | 11.0 | 7832 | 0.4641 |
|
| 66 |
+
| 0.4918 | 12.0 | 8544 | 0.4647 |
|
| 67 |
+
| 0.4933 | 13.0 | 9256 | 0.4622 |
|
| 68 |
+
| 0.499 | 14.0 | 9968 | 0.4619 |
|
| 69 |
+
| 0.4906 | 15.0 | 10680 | 0.4608 |
|
| 70 |
+
| 0.4884 | 16.0 | 11392 | 0.4622 |
|
| 71 |
+
| 0.4847 | 17.0 | 12104 | 0.4616 |
|
| 72 |
+
| 0.4916 | 18.0 | 12816 | 0.4592 |
|
| 73 |
+
| 0.4845 | 19.0 | 13528 | 0.4600 |
|
| 74 |
+
| 0.4788 | 20.0 | 14240 | 0.4594 |
|
| 75 |
+
| 0.4746 | 21.0 | 14952 | 0.4607 |
|
| 76 |
+
| 0.4875 | 22.0 | 15664 | 0.4615 |
|
| 77 |
+
| 0.4831 | 23.0 | 16376 | 0.4597 |
|
| 78 |
+
| 0.4798 | 24.0 | 17088 | 0.4595 |
|
| 79 |
+
| 0.4727 | 25.0 | 17800 | 0.4592 |
|
| 80 |
+
| 0.4736 | 26.0 | 18512 | 0.4598 |
|
| 81 |
+
| 0.4746 | 27.0 | 19224 | 0.4608 |
|
| 82 |
+
| 0.4728 | 28.0 | 19936 | 0.4589 |
|
| 83 |
+
| 0.4771 | 29.0 | 20648 | 0.4593 |
|
| 84 |
+
| 0.4743 | 30.0 | 21360 | 0.4588 |
|
| 85 |
+
| 0.4785 | 31.0 | 22072 | 0.4601 |
|
| 86 |
+
| 0.4757 | 32.0 | 22784 | 0.4597 |
|
| 87 |
+
| 0.4731 | 33.0 | 23496 | 0.4598 |
|
| 88 |
+
| 0.4746 | 34.0 | 24208 | 0.4593 |
|
| 89 |
+
| 0.4715 | 35.0 | 24920 | 0.4599 |
|
| 90 |
+
| 0.4769 | 36.0 | 25632 | 0.4622 |
|
| 91 |
+
| 0.4778 | 37.0 | 26344 | 0.4605 |
|
| 92 |
+
| 0.4798 | 38.0 | 27056 | 0.4594 |
|
| 93 |
+
| 0.4694 | 39.0 | 27768 | 0.4607 |
|
| 94 |
+
| 0.468 | 40.0 | 28480 | 0.4600 |
|
| 95 |
|
| 96 |
|
| 97 |
### Framework versions
|