Upload folder using huggingface_hub
Browse files
README.md
CHANGED
|
@@ -27,7 +27,7 @@ This model combines the SpeechBrain ECAPA-TDNN speaker embedding model with an S
|
|
| 27 |
- TIMIT test set: 6.02 cm Mean Absolute Error (MAE)
|
| 28 |
|
| 29 |
## Training Data
|
| 30 |
-
The model was trained on VoxCeleb2 dataset:
|
| 31 |
- Audio preprocessing:
|
| 32 |
- Converted to WAV format, single channel, 16kHz sampling rate, 256 kp/s bitrate
|
| 33 |
- Applied SileroVAD for voice activity detection, taking the first voiced segment
|
|
@@ -43,7 +43,7 @@ pip install git+https://github.com/griko/voice-height-regression.git
|
|
| 43 |
## Usage
|
| 44 |
|
| 45 |
```python
|
| 46 |
-
from
|
| 47 |
|
| 48 |
# Load the pipeline
|
| 49 |
regressor = HeightRegressionPipeline.from_pretrained(
|
|
|
|
| 27 |
- TIMIT test set: 6.02 cm Mean Absolute Error (MAE)
|
| 28 |
|
| 29 |
## Training Data
|
| 30 |
+
The model was trained on height enriched VoxCeleb2 dataset (for details read the paper):
|
| 31 |
- Audio preprocessing:
|
| 32 |
- Converted to WAV format, single channel, 16kHz sampling rate, 256 kp/s bitrate
|
| 33 |
- Applied SileroVAD for voice activity detection, taking the first voiced segment
|
|
|
|
| 43 |
## Usage
|
| 44 |
|
| 45 |
```python
|
| 46 |
+
from voice_height_regressor import HeightRegressionPipeline
|
| 47 |
|
| 48 |
# Load the pipeline
|
| 49 |
regressor = HeightRegressionPipeline.from_pretrained(
|