Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
language: so
|
| 4 |
+
tags:
|
| 5 |
+
- tts
|
| 6 |
+
- speaker-embedding
|
| 7 |
+
- somali
|
| 8 |
+
- speechbrain
|
| 9 |
+
- vits
|
| 10 |
+
- speecht5
|
| 11 |
+
library_name: speechbrain
|
| 12 |
+
base_model: speechbrain/spkrec-ecapa-voxceleb
|
| 13 |
+
pipeline_tag: feature-extraction
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# Ali Speaker Embedding Dataset
|
| 17 |
+
|
| 18 |
+
This dataset contains a PyTorch `.pt` file that represents a speaker embedding for the Somali male speaker **Ali**.
|
| 19 |
+
|
| 20 |
+
The embedding was generated using the [`speechbrain/spkrec-ecapa-voxceleb`](https://huggingface.co/speechbrain/spkrec-ecapa-voxceleb) speaker recognition model from over 300 audio clips of the speaker's voice.
|
| 21 |
+
|
| 22 |
+
## Dataset Details
|
| 23 |
+
|
| 24 |
+
- **File**: `Ali_speaker_embedding.pt`
|
| 25 |
+
- **Format**: PyTorch tensor (`.pt`)
|
| 26 |
+
- **Embedding Size**: 192-dimensional
|
| 27 |
+
- **Language**: Somali (`so`)
|
| 28 |
+
- **Gender**: Male
|
| 29 |
+
- **Audio Source**: 300 high-quality `.wav` files from speaker Ali
|
| 30 |
+
- **Sample Rate**: 16kHz
|
| 31 |
+
|
| 32 |
+
## Usage Example
|
| 33 |
+
|
| 34 |
+
```python
|
| 35 |
+
import torch
|
| 36 |
+
|
| 37 |
+
# Load the embedding
|
| 38 |
+
embedding = torch.load("Ali_speaker_embedding.pt")
|