jg-fisher / desktopassistant

Desktop assistant that uses speech recognition and gTTS to execute commands and talk back to the user.

Desktop Assistant

Siri like desktop assistant written in Python which uses google's speech-to-text library to process voice input.

• Install the dependencies in a virtual environment (using conda or virtualenv) to avoid any issues. Use either pip2 or pip3 for python2 and python3 respectively.

If you are a linux user install the say command using

sudo apt-get install gnustep-gui-runtime

pip2 install -r requirements.txt
pip3 install -r requirements.txt

• Usage

python desktopAssistant.py

Supported commands :

• Open reddit subreddit : Opens the subreddit in default browser.
• Open website xyz.com : self-explanatory
• Send email/email : Follow up questions such as recipient name, content will be asked in order.
• Tell a joke/another joke : Says a random dad joke.
• Current weather in {cityname} : Tells you the current condition and temperture
• weather forecast in {cityname} : Tells you the condition, highest and lowest temperture of the next two days
• What's up

desktopassistant's Issues

Google Assistance

Please be more specific

1. Provide the requirements file, dependencies. For eg. this page says that I need to have PyAudio in order for the microphone to work, this was never mentioned.
2. Instructions on how to run the program.
3. Comments about functions should rather be mentioned as docstrings for accessibility reasons.

Exit script?

I thought it may be important to end the script. Here is my way. Is this a good way?

> import sys

> elif 'exit program' in command:
>         talkToMe('Goodbye, I am shutting down.')
>         sys.exit()

ALSA lib pcm_dmix.c:1052:(snd_pcm_dmix_open) unable to open slave

I got the title issue on running the code

"OK Jarvis" initial command

I understand that this assistant is always listening on the background.

I think an initial signal would be useful to the script doesnt take a user's conversation with another person as a command to execute, unless an initial phrase is spoken first, just like the famous "OK Google".

Requirements.txt error

Name is not defined

When I try to run it, it gives the error "line 24, in my command
command = r.recognize_google(audio)
NameError: name 'audio' is not defined"

No sound

So everything works great but I have no sound and I'm not sure where to start any help would be great

can't speak

'say' is not recognized as an internal or external command,
operable program or batch file.

Spanish support

I suppose that this doesn't work in spanish "out-of-the-box", right?
How would one adjust the script for it to function with another language?