Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/Harsha200105/DesktopAssistant/llms.txt

Use this file to discover all available pages before exploring further.

Jarvis exposes three live voice-configuration commands that take effect immediately without restarting the assistant. Each change is applied to the running pyttsx3 engine straight away, and you are then asked whether you want to persist the new value to config.ini so it survives the next restart. All three handlers live in actions.py and are wired up in Jarvis2_4windows.py via the elif "change rate" in query / elif "change voice" in query / elif "change volume" in query branches.

​
Voice Gender

Say: "change voice to male" or "change voice to female" Jarvis splits the command on the word "to" to extract the target gender and then calls engine.setProperty('voice', ...) with the matching SAPI5 voice ID: 
def change_voice(query, take_command):
    try:
        voice = query.split('to')[-1]
        if voice == "male":
            engine.setProperty('voice', voices[0].id)
            speak("¿Do you want to keep this config?")
            if take_command() == "yes":
                config['DEFAULT']['voice'] = 'Male'
                with open('config.ini', 'w') as configfile:
                    config.write(configfile)
            else:
                pass

        elif voice == "female":
            engine.setProperty('voice', voices[1].id)
            speak("¿Do you want to keep this config?")
            answer = take_command()
            print(answer)
            if answer == "yes":
                config['DEFAULT']['voice'] = 'Female'
                with open('config.ini', 'w') as configfile:
                    config.write(configfile)
            else:
                pass
        else:
            speak("Invalid value. Please try again.")
    except Exception:
        speak("Invalid value. Please try again.")
voices[0] and voices[1] are the first two SAPI5 voices returned by engine.getProperty("voices"). On a standard Windows installation voices[0] is Microsoft David (male) and voices[1] is Microsoft Zira (female), though the exact voices available depend on your Windows language packs. After Jarvis confirms the change, say “yes” to write the new value to config.ini, or say anything else to keep the change only for the current session.

​
Speech Rate

Say: "change rate to 200" (replace 200 with your preferred value) The rate is extracted from the command string and passed directly to pyttsx3 as words per minute: 
def change_rate(query, take_command):
    try:
        rate = query.split('to')[-1]
        engine.setProperty('rate', int(rate))
        speak("¿Do you want to keep this config?")
        answer = take_command()
        if answer == "yes":
            config['DEFAULT']['rate'] = rate
            with open('config.ini', 'w') as configfile:
                config.write(configfile)
        else:
            pass
    except Exception:
        speak("Invalid value. Please try again.")
The default rate defined in config.ini is 150 words per minute. Typical comfortable listening ranges are 120–180 wpm; anything above 250 wpm can become difficult to follow. If you supply a non-integer value, the except block fires and Jarvis asks you to try again.

​
Volume

Say: "change volume to 80" (replace 80 with a value from 0 to 100) Volume is specified as an integer between 0 and 100. Internally it is divided by 100 before being passed to pyttsx3, which expects a float in the range 0.0–1.0: 
def change_volume(query, take_command):
    try:
        volume = query.split('to')[-1]
        engine.setProperty('volume', int(volume)/100)
        speak("¿Do you want to keep this config?")
        answer = take_command()
        if answer == "yes":
            config['DEFAULT']['volume'] = volume
            with open('config.ini', 'w') as configfile:
                config.write(configfile)
        else:
            pass
    except Exception:
        speak("Invalid value. Please try again.")
The default volume in config.ini is 100 (full volume). Setting it to 0 will silence Jarvis entirely — remember you can still use debug = True to interact via the keyboard if that happens.
Whenever Jarvis asks “¿Do you want to keep this config?”, responding with “yes” rewrites the entire config.ini file through Python’s configparser. Any manual formatting or comments you added to the file will be removed by the rewrite, because configparser does not preserve them. Keep a backup copy of your config.ini if you have custom comments in it.

​
Startup settings

At module load time actions.py reads config.ini and applies all three voice settings to the pyttsx3 engine before the first command is processed: 
engine = pyttsx3.init("sapi5")
voices = engine.getProperty("voices")

config = configparser.ConfigParser()
config.read('config.ini')

if config['DEFAULT']['voice'] == 'Male':
    engine.setProperty('voice', voices[0].id)
else:
    engine.setProperty('voice', voices[1].id)

try:
    engine.setProperty('rate', int(config['DEFAULT']['rate']))
    engine.setProperty('volume', int(config['DEFAULT']['volume'])/100)

except Exception:
    speak("Bad config. Setting up default values")
If either rate or volume contains a non-integer value when the module is imported, Jarvis announces "Bad config. Setting up default values" and continues with pyttsx3’s built-in defaults rather than crashing. Fix the values in config.ini and restart to resolve the warning.

Build docs developers (and LLMs) love

Get started for freeTalk to us