Dialect Detection is a project aimed at identifying the distinct dialects spoken in various regions of Gujarat, including Ahmedabad, Surat, Mahesana, Charotar, and Kathiyavad. Leveraging Machine Learning techniques, this project enables the detection of dialects through both live speech captured via microphone and pre-recorded audio files.
🎙️ Real-time Speech Detection: Detects dialects directly through microphone input, allowing for on-the-fly analysis.
🔍 Audio File Analysis: Capable of analyzing recorded audio files to determine the dialect used.
🤖 Machine Learning Model: Utilizes MFCC (Mel-Frequency Cepstral Coefficients) feature sets extracted from audio samples to classify dialects accurately.
To use the Dialect Detection system:
- Ensure that the necessary dependencies are installed.
- Run the system and choose the desired mode (real-time or file analysis).
- Provide the input audio source (microphone or audio file).
- Receive the detected dialect classification results.
- Python 3.x
- Machine Learning libraries (TensorFlow, Keras, Scikit-learn, Pandas, Numpy)
- Audio processing libraries (Librosa)
- Microphone access (if using real-time detection)
Contributions to the project are welcome! If you find any issues or have suggestions for improvements, feel free to open a pull request.