NLP-CV-Bengali_Speech_Chatbot

Description

A Bangla speech-based chatbot App developed using Flask microframework containing modules Speech-to-Text, Conversational Agent, and Text-to-Speech.

Environment

• Language: Python 3.10.6

Installation Procedure

1. Download the project from GitHub, Extract it and goto your base directory.
2. Download Virtual Environment sudo apt install python3.10-venv
3. Create a virtual environment using python3 -m venv <environment name>
4. Activate virtual environment source <environment name>/bin/activate
5. Install required files pip install -r requirements.txt
6. Install portaudio for PyAudio if not downloaded automatically sudo apt-get install portaudio19-dev
7. Execute the run.py file.
8. Visit http://127.0.0.1:5000 <i>[The installation procedure is for Linux (ubuntu)]</i>

Expalaination of the directory tree

- database            : Contains data and dumped models
- static              : audio, image, text, css, js files
- templates           : html files
- speech_to_text.py   : methods of ASR/STT module
- chatbot.py          : methods of CA module
- text-to-speech.py   : methods of TTS/Speech synthesis module
- face_verification.py: modules to open webcam and stream video
- face_encode.py      : encode train data
- views.py            : routing methods of the flask app
- run.py              : execute it to start server
- requirements.txt    : required dependencies to install

Video demonstrations

Bengali Voice Chatbot

video1272437095.mp4

Face Recognition

video1410013186.mp4

Note: To handle large file install git lfs

• Before installing git lfs in ubuntu

curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash
sudo apt-get install git-lfs
git lfs track path_to_file
git add path_to_file
git commit -m "commit"