Food-extraction-from-speech

Desciption:

Extract food and ingredient entity from audio file.

Diagram:

Folder structure:

├── .gitignore
├── Data/
│   ├── Audio
│   ├── Document
│   ├── Entity
│   └── Source/
│       └── Source.txt
├── Mwe List/
│   ├── CreateMweList.py
│   ├── CreateRankList.py
│   ├── Data/
│   │   ├── food_ranking.npy
│   │   ├── ingredient_ranking.npy
│   │   ├── mwe.npy
│   │   └── Source.txt
│   ├── Food_ranking.py
│   ├── Ingredient_ranking.py
│   └── MweCreation.py
├── Ner.py
├── README.md
├── requirements.txt
├── TranscribeText.py
├── tree.py
└── Utils.py

Installation:

Update pip:

    pip install --upgrade pip

Create and start virtual environments:

    pip install virtualenv
    python -m venv env
    
    # Windows activate
    env\Scripts\activate
    deactivate

    # Linux activate
    source env/bin/activate
    deactivate

Installing required packages:

    pip install -r requirements.txt

DownLoad data

    python Utils.py download_data

Usage

From command line

    $ # Speech to text 
    $ python TranscribeText.py Data\Audio\sample.mp3
    In a heavy 2-quart saucepan, mix brown sugar and nuts.
        
    $ # Entity extraction
    $ python Ner.py Data\Document\sample.txt
    ['saucepan', 'brown sugar', 'nuts']