8000 GitHub - lulrai/assistant-gpt: A fun speech-to-text and text-to-speech based assistant for your needs.
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

A fun speech-to-text and text-to-speech based assistant for your needs.

License

Notifications You must be signed in to change notification settings

lulrai/assistant-gpt

Repository files navigation

Contributors Forks Stargazers Issues

LinkedIn


AssistantGPT

A fun speech-to-text and text-to-speech based assistant for your needs.
Report Bug | Request Features

Table of Contents
  1. About The Project
  2. Features
  3. Usage
  4. Roadmap
  5. Contact
  6. Acknowledgments

About The Project

Application ScreenShot

Assistant GPT (APT) is a Python application that uses speech recognition to allow users to interact with their computer using voice commands. It supports multiple recognition engines and provides a variety of built-in commands as well as the ability to define custom commands.

(back to top)

Built With

The project was built with Python >= 3.8. All the dependencies are listed below:

  • Conda
  • Python
  • Google
  • RevChatGPT
  • PyTorch
  • SpeechRecognition
  • Coqui-Ai-TTS
  • Open-Ai-Whisper
  • PocketSphinx
  • Wit.Ai

(back to top)

Features

Always expanding features but here are a few of the current features present in the application.

  • Support for multiple speech recognition engines, including Google, Sphinx, Whisper, and Wit.ai.
  • Calibrate microphone to adjust for ambient noise before speech recognition.
  • Background listener for voice-activated commands with phrase time limit.
  • Foreground listener for speech recognition with phrase time limit.
  • Custom command creation with customizable voice triggers and actions.
  • Text-to-speech (TTS) engine integration for computer responses.
  • Interactive mode for continuous conversation with the computer.
  • (Optional) Automatic stop time for background listener.
  • Currently implemented commands:
    • TimeCommand:

      Triggers: time

      Ask the current time which will reply both current time and date.

    • ChatGPTCommand:

      Triggers: chatgpt, gpt, ask

      Ask ChatGPT questions and it will reply back.

    • ExitCommand:

      Triggers: goodbye, goodnight, abracadabra, abrakadabra

      Stops listening to further commands and terminates the program.

    • (Experimental) MessageCommand: (NOTE: ONLY WORKS ON MACOS)

      Triggers: message, text, sms, send

      Sends a message using the iMessage app on MacOS to the person you ask.

Usage

Coming soon!

(back to top)

Roadmap

  • Add further support for more text-to-speech and speech-to-text engines
  • Add more commands such as:
    • Opening applications
    • Controlling keyboard using pyautogui
    • And more!
  • Automate retrieving access_token for ChatGPT using user login.
  • Automatically generate .env variables but keep it empty.
  • Make this project a distributable library/package?

See the open issues for a full list of known issues.

(back to top)

Contact

Project Link: https://github.com/lulrai/assistant-gpt

(back to top)

Acknowledgments

(back to top)

About

A fun speech-to-text and text-to-speech based assistant for your needs.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

0