polyglot

What it is and what it does

Polyglot is a multilingual natural language processing library that wraps language models and linguistic algorithms to perform common NLP tasks across many languages. It provides a unified interface for tokenization, language detection, named entity recognition, part-of-speech tagging, sentiment analysis, word embeddings, morphological analysis, and transliteration. The package supports between 16 and 196 languages depending on the task.

The library works by loading pre-trained models for each language and task, then exposing them through a simple Python API. You instantiate a Text object with a string, and then access properties like `.language`, `.words`, `.sentences`, `.pos_tags`, `.entities`, and `.polarity` to retrieve annotations. Individual words can be examined for embeddings, morphemes, and transliteration. However, the package has not been actively maintained since 2016-07-03, and installation requires downloading external model data, which may present compatibility or availability challenges.

Use it for:

• Detect the language of user-submitted text to route it to language-specific processing pipelines.
• Extract named entities (people, locations, organizations) from multilingual documents for information extraction.
• Perform sentiment analysis on social media posts or reviews in languages other than English.
• Tokenize and tag parts of speech in non-English text for downstream linguistic analysis or machine learning.
• Transliterate text between writing systems (e.g., English to Cyrillic) for cross-script applications.
• Generate word embeddings for multilingual semantic similarity or clustering tasks.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Polyglot is a multilingual natural language processing pipeline that performs tokenization, language detection, named entity recognition, part-of-speech tagging, sentiment analysis, word embeddings, morphological analysis, and transliteration across many languages.

No—not for new projects. The package is dormant (last release 2016-07-03), classified as Beta, and carries high install friction due to external model dependencies. Maintenance status is unclear, and compatibility with modern Python versions is unverified. For current multilingual NLP work, consider actively maintained alternatives.

Install

polyglot on PyPI

pip install polyglot

uv add polyglot

poetry add polyglot

Installing polyglot

Before you install

Installation friction is high; the package has been dormant since 2016-07-03 and carries no runtime dependencies, suggesting it may require manual model downloads or system-level setup to function. Beta status and age raise questions about compatibility with current Python versions.

License in practice

GPLv3 is a copyleft license; any derivative work or modification must also be licensed under GPLv3, and distribution of the package in a proprietary application may require source disclosure or licensing negotiation.

Quickstart

import polyglot
from polyglot.text import Text, Word

text = Text("Bonjour, Mesdames.")
print(text.language.code, text.language.name)
print(text.words)
print(text.pos_tags)

Model data must be downloaded separately; the package itself has no runtime dependencies but requires external language models to function.

Verify before relying

• Whether model downloads work with current mirror or if Stony Brook DSL lab mirror is still active.
• Python 3.4+ compatibility status; classifiers list Python 3.4 but package is dormant since 2016.
• Whether high install friction is due to compiled dependencies or model data requirements.
• Current maintenance status and whether the GitHub repository accepts issues or pull requests.

Package facts

Evidence: polyglot-16.7.4.tar.gz

Keywords: polyglot

Tags

More Artificial Intelligence packages