Eyob-Sol's picture
Update README.md
88a1d5d verified
---
title: FutureCafe Voice Core
emoji: ☎️
colorFrom: indigo
colorTo: blue
sdk: gradio
app_file: app.py
pinned: false
license: mit
---
# ☎️ FutureCafe Voice & Chat Assistant
Welcome to **FutureCafe Voice & Chat Assistant** — an AI-powered demo that lets you interact with a virtual cafe agent using either **voice calls** or **chat messages**.
The assistant can answer questions about the cafe, help place food orders, confirm reservations, and provide opening hours or location info — all through a simple web interface.
---
## 🎯 Aim of the Project
The goal of this Space is to demonstrate how modern AI components (speech recognition, language models, text-to-speech) can be combined into a **realistic customer service experience** for restaurants and cafes.
FutureCafe Assistant acts like a friendly staff member:
- Answers menu or dietary questions.
- Helps with table reservations.
- Supports order placement and price calculation.
- Provides hours, address, and contact info.
- Handles both **chat** and **voice calls**.
---
## 🚀 How to Use
No installation or setup required — everything runs in the browser.
1. **Voice Call (left panel):**
- Press **Record**, speak your request, then stop recording.
- The assistant transcribes your voice, replies with text, and speaks the response back.
2. **Chat / SMS (right panel):**
- Type a message in the textbox and press Enter.
- The assistant replies in the chat window.
That’s it! 🎉 You can switch freely between chat and voice.
---
## 🛠️ Tools & Technologies
This demo integrates several lightweight but powerful AI tools:
- **[Gradio](https://gradio.app/)** – User interface for voice and chat.
- **Automatic Speech Recognition (ASR)** – Converts microphone input to text using [faster-whisper](https://github.com/SYSTRAN/faster-whisper).
- **Large Language Model (LLM)** – Core conversational logic (via [OpenAI API](https://openai.com) in this build).
- **Text-to-Speech (TTS)** – Synthesizes assistant replies into natural voice using [Piper](https://github.com/rhasspy/piper) or system TTS.
- **Python & Hugging Face Spaces** – Deployment environment.
---
## 👩‍💻 Credits
Developed as part of a portfolio project to explore **multimodal AI assistants** that combine speech, text, and reasoning for practical real-world scenarios.