crispinprojects/klatt-synthesizer
Klatt speech synthesizer
This project offers an educational example of a Klatt speech synthesizer, a system that creates spoken words by modeling the human vocal tract. It takes a date (like "Monday 1st January") and outputs an audio file of that date being spoken. This tool is for students or researchers in speech synthesis who want to understand the foundational principles of formant synthesis.
Use this if you are a student or educator wanting to learn or teach the fundamental concepts of formant speech synthesis, particularly the Klatt model, through a simple, from-scratch implementation.
Not ideal if you need high-quality, natural-sounding speech output for practical applications, as the current audio quality is described as very poor.
Stars
8
Forks
—
Language
C
License
GPL-3.0
Category
Last pushed
Feb 17, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/crispinprojects/klatt-synthesizer"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
google/voice-builder
An opensource text-to-speech (TTS) voice building tool
thuhcsi/Crystal
Crystal - C++ implementation of a unified framework for multilingual TTS synthesis engine with...
gooofy/py-picotts
Python wrappers around SVOX Pico TTS
sevangelatos/py-ttspico
Python svox picotts wrapper
LynxLine/qtspeech
QtSpeech is cross-platform library based on Qt to provide common cross-platform API to access...