Humaware VAD

🚀 [WIP]HumAware-VAD: Humming-Aware Voice Activity Detection

HumAware-VAD is a fine - tuned model based on Silero-VAD. It can distinguish humming from actual speech. Standard VAD models often misclassify humming as speech, but HumAware-VAD uses a custom dataset to improve speech detection accuracy in the presence of humming.

🚀 Quick Start

Install Dependencies

pip install torch torchaudio

Load the Model

import torch

def load_humaware_vad(model_path="humaware_vad.jit"):
    model = torch.jit.load(model_path)
    model.eval()
    return model

vad_model = load_humaware_vad()

Run Inference

import torchaudio

waveform, sample_rate = torchaudio.load("data/0000.wav")
out = vad_model(waveform)
print("VAD Output:", out)

✨ Features

• Reduce false positives where humming is mistakenly detected as speech.
• Enhance speech segmentation accuracy in real - world applications.
• Improve VAD performance for tasks involving music, background noise, and vocal sounds.

📚 Documentation

🗂️ Model Details

📄 License

This project is under the MIT license.

📄 Citation

If you use this model, please cite it accordingly.

@model{HumAwareVAD2025,
  author = {Sourabh Saini},
  title = {HumAware-VAD: Humming-Aware Voice Activity Detection},
  year = {2025},
  publisher = {Hugging Face},
  url = {https://huggingface.co/CuriousMonkey7/HumAware-VAD}
}