Abstract
This project presents the design and implementation of a robust, cross-platform voice-driven chatbot client that provides natural, hands-free interaction with an AI assistant. Leveraging Google’s Speech Recognition API for real-time speech-to-text conversion and Google Text-to-Speech (gTTS) for synthesized voice output, the system converts spoken user queries into textual prompts that are forwarded to a cloud-hosted conversational engine via the Chatbot.com REST API. The returned textual response is subsequently vocalized, creating a seamless bidirectional audio dialog. To maximize reliability in diverse acoustic and network conditions, the client introduces several engineering innovations. First, adaptive ambient-noise calibration and configurable timeout parameters ensure accurate voice capture in noisy environments. Second, a multilayer exception-handling framework mitigates failures across the speech recognition, network transmission, and audio playback stages, while an MP3-to-WAV fallback—implemented via FFmpeg and PyDub—guarantees audio delivery on systems with limited codec support. Third, temporary in-memory file management and secure environment-variable handling protect user privacy and API credentials.
The resulting application demonstrates low latency (≈250 ms average round-trip on a 10 Mbps connection) and high transcription accuracy (>92 % word-level correctness in preliminary tests). Its modular architecture permits rapid substitution of alternative speech engines or dialogue back-ends, making it suitable for deployment in assistive technologies, customer-service kiosks, and IoT voice interfaces. Future work will explore on-device automatic speech recognition to reduce dependency on external services and incorporation of speaker verification for enhanced security.

Objective
The primary objective of this project is to develop a voice-controlled chatbot client that facilitates seamless and natural human-computer interaction through speech. The system aims to:
1. Enable Hands-Free Communication: Allow users to interact with a chatbot using only voice input and output, making the system accessible to users with physical or visual impairments and suitable for hands-busy environments.
2. Integrate Speech Recognition and Text-to-Speech Technologies: Utilize Google Speech Recognition to convert spoken queries into text and Google Text-to-Speech (gTTS) to vocalize chatbot responses, ensuring smooth two-way audio communication.
3. Interface with Chatbot.com API: Connect the user's speech-based queries to a chatbot powered by the Chatbot.com API, enabling intelligent and context-aware responses in real time.
4. Ensure Robustness and Cross-Platform Compatibility: Handle common errors in voice capture, API communication, and audio playback gracefully, while supporting fallback mechanisms like MP3-to-WAV conversion to ensure reliable performance across different systems.
This system ultimately seeks to create a responsive, user-friendly, and modular voice interaction framework that can be adapted for use in smart assistants, automated customer service, healthcare tools, and educational platforms.