CV Parser (Vietnamese + English)

This project extracts structured information from Vietnamese and English CVs (PDF or DOCX), including name, contact, skills, education, experience, and languages. It auto-detects the language and routes outputs accordingly.

---

📦 External Requirements

Please install these system-level dependencies before running the project:

Dependency	Version	Description
Python	3.12.7+	Recommended interpreter
Poppler	24.08.0	Required for accurate PDF text extraction
Tesseract-OCR	Latest / 5.x+	OCR fallback for scanned/non-text PDFs
PyTorch	Compatible w/ Transformers	Required for spaCy's transformer pipeline
Transformers	Latest	Used for spaCy's en_core_web_trf model

Linux (Debian/Ubuntu)

sudo apt update
sudo apt install poppler-utils tesseract-ocr

MacOS (Homebrew)

brew install poppler tesseract

Python Setup

1. Clone the repo

git clone https://github.com/Watch650/ResumeParser.git

2. Create a virtual environment

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

3. Install Python dependencies

pip install -r requirements.txt

4. Download spaCy models

python -m spacy download en_core_web_trf

Folder Structure

cv-parser/
├── CV/                     # Drop .pdf/.docx files here
├── text_extract/           # Output: Cleaned text per language
├── parsed_data/            # Output: Structured JSON data
├── extractors/             # Text extractors (PDF, DOCX)
├── utils/                  # Helper utilities (cleaning, OCR)
├── file_router.py          # Routes and processes input CV files
├── file_parser_en.py       # English CV parser
└── file_parser_vn.py       # Vietnamese CV parser

How to Use

1. Drops files

Place .pdf or .docx files in the CV/ folder.

2. Extract & route text

python file_router.py

3. Parse CVs

python file_parser_en.py   # For English CVs
python file_parser_vn.py   # For Vietnamese CVs

Ouput Format

Each parsed CV will be saved to: parsed_data/extracted_cv_data.json. Sample entry:

{
  "ho_ten": "Vu Hoang Lan",
  "email": "example@gmail.com",
  "so_dien_thoai": "+84987654321",
  "hoc_van": [...],
  "kinh_nghiem": [...],
  "ky_nang": [...],
  "ngoai_ngu": [...],
  "source_file": "CV_1_extracted_text.txt"
}

Logging

Logs are saved in: logs/