AI Based Voice Assistant

Author

• Shivam Singh Sikarwar
• Department of Electronics and Communication, MITS Gwalior

Abstract

• Digital installations simplify life using auxiliary technology tools.
• Examples: Amazon Alexa, Apple Siri, Microsoft Cortana.
• Voice assistant: software using AI to take voice input and perform tasks.
• Converts speech to text (STT) and text to speech (TTS).
• Divided into branches: computer-related environment, personal interaction, and information systems.
• Uses Python packages; focuses on NLP for contextual text recognition.
• Works online and offline; data stored internally to reduce complexity.

Introduction

• Voice assistants run computer systems via verbal commands.
• Perform functions: email, to-do lists, opening websites/apps, sending texts.
• Offer convenience and speed, reducing the need for typing.
• Requires Python knowledge; uses libraries like speech recognition, pyttsx3.

Literature Review

• Historical Development
  • 1952: Bell Labs' 'Audrey' - understood 10 digits.
  • 1960s: IBM's shoebox machine - understood 16 words.
  • Hidden Markov Model (HMM) improved word recognition.
  • Introduction of Siri by Apple.
• Modern Voice Assistants
  • Enhance interaction between humans and machines.
  • Convert verbal commands into actions using algorithms.

Proposed Methodology

• Voice assistants interpret voice inputs to perform tasks using Python.
• Supports visually impaired users.
• Utilizes Python libraries: speech recognition, pyttsx3, Python backend, system calls.

Result and Discussion

• Efficient and quick task execution using system's microphone.
• Always accessible, performs tasks based on user needs.
• Demonstrations include opening Wikipedia and Spotify.

Conclusion and Future Work

• Voice assistants simplify daily tasks, using Python and AI.
• Work across various OS, with high efficiency.
• Automation reduces reliance on typing.
• Future improvements with IoT technology for enhanced efficiency and accuracy.

References

1. Douglas O'Shaughnessy, IEEE
2. Nil Goksel-Canbek & Mehmet Emin Mutlu
3. Information on Cortana, Google Assistant, Apple Siri
4. Various online resources on AI and Python
5. Python programming tutorials and guides