ArabicVoiceSimilarity is a Python-based project that combines speech recognition and natural language processing to transcribe Arabic audio input and evaluate the similarity between the transcribed text and a predefined correct answer. This project leverages the power of the SpeechRecognition library for audio transcription and the SentenceTransformer library for computing semantic similarity.
- Real-time Speech Recognition: Capture and transcribe Arabic speech using a microphone.
- Semantic Similarity Evaluation: Compare the transcribed text with a predefined correct answer to determine similarity.
- Multilingual Model Support: Utilizes a pre-trained multilingual model to handle diverse linguistic nuances.
- Threshold-based Validation: Determine the correctness of the transcribed answer based on a configurable similarity threshold.
- Python 3.x
-
Clone the repository:
git clone https://github.com/yourusername/ArabicVoiceSimilarity.git cd ArabicVoiceSimilarity -
Install the required packages using
requirements.txt:pip install -r requirements.txt
- Ensure your microphone is set up and working.
- Run the script to start transcribing audio input.
- Speak into the microphone when prompted.
- The script will transcribe your speech and compute the similarity score with the correct answer.
- Based on the similarity score, the script will indicate whether the answer is correct.
Contributions are welcome! Please feel free to submit a pull request or open an issue to discuss improvements or features.
This project is licensed under the GPL-3.0 license.