Simple system that makes usage system via voice available

Shkvaldev 08a92b29bd little README improvements 1 year ago
API ab0f00dfb9 first commit 1 year ago
Modules ab0f00dfb9 first commit 1 year ago
ModulesParser ab0f00dfb9 first commit 1 year ago
Recognizer ab0f00dfb9 first commit 1 year ago
Speaker ab0f00dfb9 first commit 1 year ago
TGBot ab0f00dfb9 first commit 1 year ago
README.md 08a92b29bd little README improvements 1 year ago
config.json ab0f00dfb9 first commit 1 year ago
main.py ab0f00dfb9 first commit 1 year ago
packages.txt 08a92b29bd little README improvements 1 year ago
requirements.txt ab0f00dfb9 first commit 1 year ago

README.md

COMA

Open source system that tries to implement useful voice UI

This software is under development!

All start modules are shipped in russian by default!

Targets

[*] Main target

  • Making it possible for blind or visually impaired people to use computers without any obstacles

Important note: this software could not exist without the following components

  • Vosk by Alphacephei (is used to recognize user speech in server-side software)
  • RHVoice by Olga Yakovleva (is used to generate answers for user in server-side software)

Installation

There is only Arch Linux support for now!

  1. Clone the repository: shell git clone (repo url here) cd COMA
  2. Install all system dependences: shell sudo pacman -Sy $(cat packages.txt)
  3. Install all python dependences: shell pip install -r requirements.txt
  4. Run the server: shell python main.py

Platforms

  • Linux (server)
  • All OSs that have Telegram support (client)

TODO

  • Popular initialization systems support (SystemD, OpenRC, RunIt)
  • Simple installator
  • Simple CLI or WEB manager
  • Docker support
  • Independent clients for Linux and Android
  • Anvanced system for creating conversations
  • Extend API