YannickJadoul/Parselmouth

What it solves

Parselmouth bridges the gap between the specialized phonetic analysis software Praat and the Python ecosystem. It allows researchers and developers to use Praat's powerful speech processing algorithms directly within Python scripts, avoiding the need to rely on Praat's own scripting language or manual interface.

How it works

Unlike other tools that wrap Praat's scripting language or attempt to recreate its features, Parselmouth provides a direct interface to Praat's underlying C/C++ code. This ensures that the algorithms and their outputs are identical to those produced by the original software. It exposes this functionality through a standard, Pythonic interface, allowing users to manipulate audio data and perform analyses using familiar libraries like NumPy and Matplotlib.

Who it’s for

It is designed for phoneticians, speech scientists, and researchers who need to perform large-scale or automated audio analysis while leveraging the established, reliable algorithms found in Praat.

Highlights

• Direct access to Praat's core C/C++ algorithms for consistent results.
• Pythonic API that integrates seamlessly with standard data science tools.
• Cross-platform support (Linux, macOS, and Windows) via standard Python package installation.
• Enables automation of complex phonetic analysis workflows across large datasets.