Palavras PT-BR is an open-source collection of 1,916,706 Brazilian Portuguese words. It provides the vocabulary both as a plain-text file—with one word per line—and as a ready-to-query SQLite database.

The project is useful for spell checkers, word games, search tools, text processing, linguistic experiments, autocomplete systems, and any application that needs a broad Portuguese vocabulary without building a dataset from scratch.

Available formats

• Palavras_PT-BR.txt: a simple text file containing one word per line.
• Palavras_PT-BR.db: a SQLite database ready for structured searches and integrations.
• Online preview: a browser interface for searching and exploring words in the database.

How the collection was built

The initial vocabulary came from the IntelliJ dictionary. It was expanded with Brazilian Portuguese verbs collected from public conjugation resources, while spaCy helped identify infinitive forms before the verbs were conjugated across different tenses. Roman numerals and suspicious entries were removed, and missing words were also added manually.

Included Python scripts

• db_create.py: recreates the SQLite database.
• db_insert.py: inserts words into the database.
• db_select.py: queries stored words.
• db_delete.py: removes words from the database.
• AVLTree.py: implements an AVL binary search tree.
• example_usage_tree.py: demonstrates how to use the AVL tree.

Try it online

You can explore the browser-based preview at Palavras PT-BR Preview .

The complete dataset and source code are available on GitHub - Palavras PT-BR .

CMD / Terminal
git clone https://github.com/AlfredoFilho/Palavras_PT-BR.git