Update README.md
Browse files
README.md
CHANGED
@@ -22,4 +22,32 @@ configs:
|
|
22 |
data_files:
|
23 |
- split: generated
|
24 |
path: data/generated-*
|
|
|
|
|
|
|
25 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
data_files:
|
23 |
- split: generated
|
24 |
path: data/generated-*
|
25 |
+
license: cc0-1.0
|
26 |
+
language:
|
27 |
+
- hy
|
28 |
---
|
29 |
+
|
30 |
+
# Hy-Generated Audio Data 2
|
31 |
+
|
32 |
+
This dataset provides Armenian speech data consisting of generated audio clips and is addition to [this](https://huggingface.co/datasets/ErikMkrtchyan/Hy-Generated-audio-data-with-cv20.0) dataset.
|
33 |
+
|
34 |
+
- The `generated` split contains 137,419 high-quality clips synthesized using a fine-tuned [F5-TTS](https://github.com/f5-lab/tts) model, covering 404 equal distribution of synthetic voices.
|
35 |
+
|
36 |
+
---
|
37 |
+
|
38 |
+
## 📊 Dataset Statistics
|
39 |
+
|
40 |
+
| Split | # Clips | Duration (hours) |
|
41 |
+
|-------------|-----------|------------------|
|
42 |
+
| `generated` | 137,419 | 173.76 |
|
43 |
+
|
44 |
+
**Total duration:** ~**173 hours**
|
45 |
+
|
46 |
+
|
47 |
+
## 🛠️ Loading the Dataset
|
48 |
+
|
49 |
+
```python
|
50 |
+
from datasets import load_dataset
|
51 |
+
|
52 |
+
dataset = load_dataset("ErikMkrtchyan/Hy-Generated-audio-data-with-cv20.0")
|
53 |
+
```
|