🗣️ Speech-to-Speech Assistant

It's a minimalistic and efficient streamlit chatbot for Urdu speakers to ask questions and get response in both text and audio formats. It's powered by Google Gemini.

---

✨ Features

• Converts Speech to Text: Simply speak, and the app captures your input in Urdu.
• Responds with Speech: Get a spoken response back in Urdu from the LLM, ensuring seamless communication.
• Chat History Tracking: The assistant keeps track of the conversation so it feels continuous.

🌟 The best part? This entire app only took 50 lines of code!

---

📚 Libraries Used

• Streamlit: For building the interactive web interface.
• gTTS: Google Text-to-Speech for generating spoken responses.
• Google Generative AI: To power the conversational response generation.
• Streamlit Mic Recorder: For capturing speech input.

---

🛠️ Installation & Setup

Step 1: Clone the Repository

git clone https://github.com/aasherkamal216/STT_TTS.git
cd STT_TTS

Step 2: Set Up a Virtual Environment (Optional but Recommended)

python -m venv venv
venv\Scripts\activate

Step 3: Install Dependencies

pip install -r requirements.txt

Step 4: Set Up API Key

In the .streamlit/secrets.toml file, add your Google API key:

GOOGLE_API_KEY = "your_google_gemini_api_key"

---

🚀 Running the App

To start the app, simply run:

streamlit run app.py

---

This assistant app provides a fast, interactive experience with only minimal code!