phonemizer-fork

What it is and what it does

Phonemizer-fork is a Python library that converts written text into phonetic transcriptions—sequences of individual speech sounds (phones)—in multiple languages. It wraps four different phonemization backends (espeak, espeak-mbrola, festival, and segments), each with different capabilities: espeak supports IPA output for many languages at fast speed, espeak-mbrola uses SAMPA for a smaller set of languages but slower, festival handles American English with syllable-level tokenization, and segments uses user-defined grapheme-to-phoneme mappings. The library provides both a command-line tool and a Python API for straightforward text-to-phone conversion.

You would use this package when you need to convert text into phonetic form for speech synthesis, linguistic analysis, or training speech-related machine learning models. The choice of backend depends on your language, required phonetic alphabet (IPA vs. SAMPA vs. custom), and whether you need syllable-level or word-level tokenization. It handles punctuation preservation and stressed phones where supported, and integrates with joblib for parallel processing.

Use it for:

• Prepare training data for speech systems by converting text to phonetic representations using espeak or festival
• Analyze pronunciation patterns across languages for linguistic research or language learning applications
• Build custom phonemization workflows for languages using the segments backend with user-provided mappings
• Extract syllable-level phonetic tokens for prosody analysis or rhythm-based speech processing in English
• Convert multilingual text to IPA for comparative phonetic studies

Worth the install?

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

Converts text to phonetic representations (phones) in multiple languages using pluggable backends like espeak, espeak-mbrola, festival, or segments.

Yes, if you need multilingual text-to-phoneme conversion. The package is actively maintained, has low install friction, and offers flexible backend selection for different use cases. The GPLv3+ license is a hard constraint only if you're building proprietary software; for research, open-source, or internal projects it poses no barrier. No known vulnerabilities.

Install

phonemizer-fork on PyPI

pip install phonemizer-fork

uv add phonemizer-fork

poetry add phonemizer-fork

Installing phonemizer-fork

Before you install

Low install friction with a pure-Python wheel and only 5 runtime dependencies. Repository is active with recent commits and 1567 stars.

License in practice

Licensed under GPLv3+, a copyleft license requiring that any derivative works or distributions remain under the same terms. Suitable for open-source projects but incompatible with proprietary software that cannot be released under GPL.

Quickstart

pip install phonemizer-fork
from phonemizer.phonemize import phonemize
phones = phonemize('hello', language='en-us', backend='espeak')

Backend engines (espeak, festival, espeak-mbrola) may require separate system installation depending on which backend you select.

Verify before relying

• Whether backend engines are bundled or must be installed separately on the system
• Whether this fork is actively maintained or has diverged from the original phonemizer package
• Exact language and feature support for each backend beyond the comparison table

Package facts

Evidence: phonemizer_fork-3.3.2-py3-none-any.whl

Keywords: G2P, TTS, espeak, festival, linguistics, phone

Tags

More Linguistic packages