update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: mit
|
| 5 |
+
base_model: microsoft/speecht5_tts
|
| 6 |
+
tags:
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- speechbrain/spkrec-xvect-voxceleb
|
| 10 |
+
model-index:
|
| 11 |
+
- name: speecht5_finetuned_voxpopuli_en
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# speecht5_finetuned_voxpopuli_en
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the speechbrain/spkrec-xvect-voxceleb dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.4466
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 5e-06
|
| 42 |
+
- train_batch_size: 8
|
| 43 |
+
- eval_batch_size: 2
|
| 44 |
+
- seed: 42
|
| 45 |
+
- gradient_accumulation_steps: 4
|
| 46 |
+
- total_train_batch_size: 32
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: linear
|
| 49 |
+
- lr_scheduler_warmup_steps: 1500
|
| 50 |
+
- training_steps: 12000
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 56 |
+
| 0.5271 | 1.58 | 1000 | 0.4843 |
|
| 57 |
+
| 0.5025 | 3.15 | 2000 | 0.4648 |
|
| 58 |
+
| 0.4947 | 4.73 | 3000 | 0.4575 |
|
| 59 |
+
| 0.4857 | 6.3 | 4000 | 0.4543 |
|
| 60 |
+
| 0.487 | 7.88 | 5000 | 0.4519 |
|
| 61 |
+
| 0.4804 | 9.46 | 6000 | 0.4494 |
|
| 62 |
+
| 0.4798 | 11.03 | 7000 | 0.4487 |
|
| 63 |
+
| 0.4813 | 12.61 | 8000 | 0.4478 |
|
| 64 |
+
| 0.4831 | 14.18 | 9000 | 0.4476 |
|
| 65 |
+
| 0.4751 | 15.76 | 10000 | 0.4468 |
|
| 66 |
+
| 0.4795 | 17.34 | 11000 | 0.4463 |
|
| 67 |
+
| 0.474 | 18.91 | 12000 | 0.4466 |
|
| 68 |
+
|
| 69 |
+
|
| 70 |
+
### Framework versions
|
| 71 |
+
|
| 72 |
+
- Transformers 4.32.0.dev0
|
| 73 |
+
- Pytorch 1.13.1
|
| 74 |
+
- Datasets 2.13.1
|
| 75 |
+
- Tokenizers 0.13.2
|