Open Source C Speech Software

Open JTalk is a Japanese text-to-speech synthesis system. This software is released under the Modified BSD license.

eGuideDog project develops free software for the blind. Currently, we focus on WebSpeech, Ekho TTS and WebAnywhere.

Praat is a speech analysis application and source repository for doing phonetics by computer. It was created by Paul Boersma and David Weenink at the University of Amsterdam. The software lets researchers, linguists, clinicians, teachers, and students analyze, synthesize, manipulate, and annotate speech recordings. It supports acoustic inspection through waveforms, spectrograms, pitch tracks, formants, intensity, and related phonetic measurements. Praat also includes scripting tools for automating repetitive analysis, building custom workflows, and processing larger speech datasets. The project is especially useful for academic phonetics, speech science, language documentation, pronunciation research, and publication-quality speech visualization.

hts_engine is software to synthesize speech waveform from HMMs trained by the HMM-based speech synthesis system (HTS). This software is released under the Modified BSD license.

eSpeak NG is a free, open-source text-to-speech synthesizer that converts written text into spoken audio across more than 100 languages and accents. Developed as a continuation of eSpeak, it uses formant synthesis to generate speech with a compact memory footprint. The software supports adjustable voices, speaking rates, pitch, and pronunciation characteristics. It can produce spoken output in real time or export synthesized speech to WAV files. eSpeak NG also provides phoneme translation tools and supports Speech Synthesis Markup Language for controlling speech output. Its command-line interface, shared library, and Windows SAPI5 integration enable use in accessibility tools and other applications. It supports Linux, Windows, macOS, Android, and additional platforms.

SPTK is a suite of speech signal processing tools for UNIX environments, e.g., LPC analysis, PARCOR analysis, LSP analysis, PARCOR synthesis filter, LSP synthesis filter, vector quantization techniques, and other extended versions of them.

moonshine is an open-source automatic speech recognition toolkit optimized for fast and accurate transcription on edge devices and local environments. The project is designed to enable real-time voice applications such as live transcription, voice commands, and embedded speech interfaces without requiring heavy cloud infrastructure. Its architecture emphasizes low latency and flexible input handling, allowing audio streams of varying durations rather than relying on fixed processing windows. Moonshine supports multiple platforms including mobile, desktop, and embedded systems, and provides example projects to accelerate integration into real-world products. The toolkit also includes specialized model variants, including monolingual options that improve accuracy for specific languages. Overall, moonshine serves developers building privacy-conscious, on-device voice interfaces that demand high performance with minimal resource overhead.

a tool for segmenting, labeling and transcribing speech

Source code from the Research Institute for Signals, Systems and Computational Intelligence http://fich.unl.edu.ar/sinc

TeamBlibbityBlabbity is an attempt to document and provide an example implemention of the proprietary TeamSpeak 2 protocol.

The Open VXI VoiceXML interpreter is a portable open source library that interprets the VoiceXML dialog markup language. It is designed to serve as a reference for parties interested in understanding how VoiceXML markup might be executed.

Winboard 4.5 32-bit is a free Windows accessible Chess program that works automatically with the JAWS or the free NVDA screen reader. It is for the blind, low sighted or those who can not use a mouse. It provides vocal announcements of position changes and other selectable board conditions. Blind players also use a separate "tactile chess board". Winboard 4.5 has full keyboard access to move pieces and run menu items. Partial sighted players use high contrast mode and adjust board, piece, most font sizes and colors. Available languages are English, German, Spanish, Italian, Dutch and Russian. Games may be viewed, modified or saved in standard PGN format. Two Chess engines supply play and grand master strength move analysis. Connect to the Free Internet Chess Server (FICS) and play humans. Modify verbosity in screen reader configurations and in the Winboard "Sound" dialog under the "General" menu. Find all documentation in the "doc" folder under the Winboard directory of your C drive.

A patent-free audio codec designed especially for voice (unlike Vorbis which targets general audio) signals and providing good narrowband and wideband quality. This project is complementary to the Ogg Vorbis codec.

The project provides a ready-to-use interface for the julius CSR engine for a handicapped child which is not able to use the keyboard well. It integrates into X11 and Windows. Find out how you can help: http://simon-listens.org/index.php?support

To provide basic text-to-speech capability on as many platforms and for as many spoken languages as possible by formant synthesis from an International Phonetic Alphabet representation.

MyDSReader is an homebrew for the Nintendo DS that helps visually impaired users: 1. Read documents in digital format (text, word, pdf, DAISY) 2. Take voice annotations 3. Read e-mails and reply/write using recorded voice clips

Unofficial fork of the GyachE (www.phrozensmoke.com), intended for bug-fixes and further development.

VoiceCode is an Open Source initiative started by the National Research Council of Canada, to develop a programming by voice toolbox. The aim of the project is to make programming through voice input as easy and productive as with mouse and keyboard. For install, Use subversion, as described in this page: http://sourceforge.net/apps/mediawiki/voicecode/index.php?title=VCode_1_Doc/InstallationManual.

This project will show how to implement the Hidden Markov Model approximations of Voice Recognition into embedded and low power systems.

Text-to-Speech conversor for Basque and Spanish. It includes linguistic processing and built voices for the languages aforementioned. Its acoustic engine is based on hts_engine and it uses a high quality vocoder called AhoCoder. Developed by Aholab Signal Processing Laboratory: https://aholab.ehu.es/aholab/ http://aholab.ehu.es/ahocoder/

Text-to-Speech conversor for Basque, Spanish, Catalan, Galician and English. It includes linguistic processing and built voices for all the languages aforementioned. Its acoustic engine is based on hts_engine and it uses a high quality vocoder called AhoCoder. Developed by Aholab Signal Processing Laboratory: https://aholab.ehu.es/aholab/ http://aholab.ehu.es/ahocoder/

Speak Freely is a Cross Platform Internet telephony (Voice Chat) application which provides high quality voice grade audio with GSM and CELP compression and encryption with DES, Blowfish, and IDEA ciphers. Refering to recent news: The maintainers will never add any unwanted extra software in the zip download file but we can only speak for our selves.

Tiny player (up to 3 kb) for Jescola's Buzz. Includes Klatt speech synth (experimental). First release in 64k intro Yo-08 which took first place on Paradox 2002. C-version was created to be ported to assembler.

Dhvani is Text-to-Speech System for Indic Languages. Current C- GNU/Linux implementation supports Hindi, Kannada, Marathi, Malayalam, Gujarati, Bengali, Telugu, Panjabi, Tamil and Oriya.

Scalable Language API (SLAPI) The most comprehensive architecture for conversational natural-language applications including speech recognition/synthesis, semantics, & machine translation. Used on Android & other mobile app platforms.