A Digital Sciences for Hi-Tech final project by TUKI:
- Sahar Machavy
- Omri Shmueli
And guided by Martin Raskin
Using voice detection, text-to-speech (gTTS), a large language model (PaLM API), and a transcription service (Faster-Whisper), we have created a holistic experience to improve English speaking.
The product aspires to serve those who seek to improve their spoken language skills with a conversation-based approach. It also includes a testing module that grades the user based on their fluency, grammar, and ability to provide certain information.
- Clone this repository.
- Install Python and a IDE (VS Code or PyCharm).
- Install MySQL from the following link - Link
a. Follow the installation process.
b. Use "root" user and define its password to be "password123" or any password of your choice (required to be configured later in the configuration.json file) - Create a dedicated virtual environment and install the requirements from the .txt file using
python pip install -r requirements.txt - Set an environment variable named "PALM_API_KEY" with its value set to your PaLM API key taken from here- Link
- Reboot your system.
A user guide can be found in the repository wiki - Link to user guide