# Audio Analysis with D-Vectors
This is a Streamlit application that compares two audio samples using d-vectors. It utilizes the Pinecone similarity search service for finding the nearest speaker based on the computed d-vectors.
## Prerequisites
- Python 3.7 or higher
## Installation
1. Clone the repository:
```shell
git clone https://github.com/sujanMidatani7/d_vector.git
- Navigate to the project directory:
cd d_vector
- Install the required dependencies:
pip install -r requirements.txt
-
Make sure you have obtained a Pinecone API key. If not, sign up for a Pinecone account and create an API key.
-
Open the
dVectorSA.py
file and replace'f9571b23-70be-4556-893a-7342b0bb51d1'
in thepinecone.init()
function with your Pinecone API key. -
Run the Streamlit app:
streamlit run dVectorSA.py
-
The application will open in your browser. You can select the first audio file using the file uploader component.
-
After selecting the audio file, the app will compute the d-vector for the audio file and find the nearest speaker using the Pinecone index.
-
The app will display the nearest speaker's ID and score.
Contributions are welcome! If you find any issues or have suggestions for improvement, please open an issue or submit a pull request.
This project is licensed under the GNU License.