End of training
Browse files
README.md
CHANGED
@@ -16,6 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
# speecht5_tts_sinhala
|
17 |
|
18 |
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the sinhala_tts dataset.
|
|
|
|
|
19 |
|
20 |
## Model description
|
21 |
|
@@ -35,15 +37,26 @@ More information needed
|
|
35 |
|
36 |
The following hyperparameters were used during training:
|
37 |
- learning_rate: 1e-05
|
38 |
-
- train_batch_size:
|
39 |
- eval_batch_size: 8
|
40 |
- seed: 42
|
41 |
- gradient_accumulation_steps: 2
|
42 |
-
- total_train_batch_size:
|
43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
- lr_scheduler_type: linear
|
45 |
- lr_scheduler_warmup_steps: 500
|
46 |
- training_steps: 4000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
47 |
|
48 |
### Framework versions
|
49 |
|
|
|
16 |
# speecht5_tts_sinhala
|
17 |
|
18 |
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the sinhala_tts dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.4122
|
21 |
|
22 |
## Model description
|
23 |
|
|
|
37 |
|
38 |
The following hyperparameters were used during training:
|
39 |
- learning_rate: 1e-05
|
40 |
+
- train_batch_size: 16
|
41 |
- eval_batch_size: 8
|
42 |
- seed: 42
|
43 |
- gradient_accumulation_steps: 2
|
44 |
+
- total_train_batch_size: 32
|
45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
46 |
- lr_scheduler_type: linear
|
47 |
- lr_scheduler_warmup_steps: 500
|
48 |
- training_steps: 4000
|
49 |
+
- mixed_precision_training: Native AMP
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
54 |
+
|:-------------:|:------:|:----:|:---------------:|
|
55 |
+
| 0.4719 | 28.17 | 1000 | 0.4329 |
|
56 |
+
| 0.4419 | 56.34 | 2000 | 0.4149 |
|
57 |
+
| 0.4281 | 84.51 | 3000 | 0.4132 |
|
58 |
+
| 0.4358 | 112.68 | 4000 | 0.4122 |
|
59 |
+
|
60 |
|
61 |
### Framework versions
|
62 |
|