Analysis updated 2026-05-18
Study a working example of wake word detection combined with speech to text and text to speech.
Use as a starting template for building your own voice-controlled desktop assistant.
Trigger simple voice commands like opening a website, hearing a joke, or checking the time.
| b-suhas/tuntun-mausi-voice-assistant | 0xhassaan/nn-from-scratch | 3ks/embedoc | |
|---|---|---|---|
| Stars | 0 | 0 | — |
| Language | Python | Python | Python |
| Last pushed | — | — | 2023-06-08 |
| Maintenance | — | — | Dormant |
| Setup difficulty | easy | moderate | hard |
| Complexity | 2/5 | 4/5 | 1/5 |
| Audience | vibe coder | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Needs a microphone and an internet connection for Google speech recognition, Linux users must install extra audio packages.
Tuntun Mausi is a wake word voice assistant built in Python as a beginner learning project. It listens in the background and only springs into action once it hears the phrase "Tuntun Mausi". After that it waits for a spoken command and answers out loud in Hinglish, a casual mix of Hindi and English common in everyday conversation in India, giving it a playful and informal personality rather than a formal, robotic tone. Once woken up, the assistant can open websites such as Google, YouTube, LinkedIn, and GitHub in your browser, run a Google search based on a spoken query, tell you the current time or date, and speak a random joke. Casual small talk is also built in, so greetings like hello or thank you get a Hinglish-flavored reply. Voice input is converted to text using Google's speech recognition service, which needs an internet connection, while the spoken replies are generated offline through a text-to-speech library. It is built from a handful of open-source Python libraries: speech_recognition for capturing and converting voice, pyttsx3 for offline text-to-speech, pyjokes for the joke feature, and Python's built-in webbrowser module for opening sites. It runs on Windows, macOS, or Linux, though Linux users may need to install a couple of extra system packages for the microphone and speech output to work correctly. The project is meant for someone learning how speech recognition, text-to-speech, and simple browser automation fit together in Python, rather than for production use. The author describes several planned improvements, including an offline wake word detector, weather updates, music playback, and reminders, none of which are implemented yet.
A beginner Python project: a Hinglish speaking voice assistant that wakes on a spoken phrase and can open websites, tell jokes, and answer casual chat.
Mainly Python. The stack also includes Python, speech_recognition, pyttsx3.
No license file is mentioned in the README, so terms of reuse are unclear.
Setup difficulty is rated easy, with roughly 30min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.