RoomFinder is a desktop application powered by machine learning and voice interaction that assists users in locating instructors and rooms within a specific building or campus area.
The system works entirely offline, meaning it does not require an internet connection to function. Users simply speak their queries, and the application processes the request using speech recognition and machine learning to determine the requested location. The result is then communicated through text-to-speech while also displaying the location visually within the application.
This project demonstrates the integration of voice interaction, machine learning, and graphical user interfaces to create a practical navigation tool.
- Python
- Kivy
- KivyMD
- Scikit-Learn
- PyAudio
- Speech Recognition
- Text-to-Speech
- Machine Learning Classification
- Voice Command Processing
- Offline Application Development
- GUI Development
-
Identified the Problem Users often struggle to find rooms or instructors in large buildings. The goal was to create a simple tool that can answer location queries.
-
Designed the Voice Interaction System Implemented speech input so users could interact with the system through natural voice commands.
-
Integrated Machine Learning Used Scikit-Learn to build a model that processes and interprets user queries.
-
Implemented Audio Processing Used PyAudio to capture microphone input and process the voice commands.
-
Added Text-to-Speech Response After identifying the requested location, the system responds with an audible answer using text-to-speech.
-
Developed the User Interface Built the graphical interface using Kivy and KivyMD to display location results visually.
-
Ensured Offline Functionality Designed the system so all processing occurs locally on the machine.
Through this project, I learned:
- How to integrate machine learning into a functional application
- How to implement voice-based user interfaces
- Handling audio input and processing in Python
- Developing cross-platform graphical interfaces with Kivy
- Designing offline AI-powered applications
This project helped improve my ability to:
- Design and implement end-to-end software projects
- Combine machine learning with user interface development
- Debug complex systems involving audio processing and ML models
- Translate a real-world problem into a practical software solution
It also strengthened my understanding of building interactive and intelligent applications.
Future improvements for this project may include:
- Implementing Natural Language Processing (NLP) for better query understanding
- Adding visual indoor maps for navigation
- Supporting multiple languages
- Improving speech recognition accuracy
- Expanding the system to support larger building databases
- Deploying the application to Android devices using Kivy
git clone https://github.com/astigPree/GuestOCAI.git
cd GuestOCAIpip install kivy kivymd scikit-learn pyaudio speechrecognition pyttsx3python main.py