Indic Multilingual Speech Bot

Overview

AI4ALPHA is a multilingual speech bot application built using Streamlit. The app enables users to record audio, transcribe speech, translate it into multiple languages, and generate responses using a language model. It supports language identification, speech-to-text transcription, and text translation.

Features

• Record and save audio
• Language identification
• Speech-to-text transcription
• Speech-to-text translation
• Text-to-text translation
• Text-to-speech translation for non-Indian languages
• Integration with a language model for generating responses

Installation

Prerequisites

• Python 3.8 or higher
• pip (Python package installer)

Step-by-Step Installation

1. Clone the repository:

   git clone https://github.com/your-username/ai4alpha.git
   cd ai4alpha
   

2. Create a virtual environment:

   python -m venv venv
   

3. Activate the virtual environment:

   • On Windows:

     venv\Scripts\activate
     

   • On macOS/Linux:

     source venv/bin/activate
     

4. Install the required packages:

   pip install -r requirements.txt
   

Running the Application

1. Ensure the virtual environment is activated.

2. Run the Streamlit app:

   streamlit run app.py
   

3. Open the app in your web browser: The app will usually be accessible at http://localhost:8501

Project Structure

Usage

1. Record Audio:

   • Click on the record button to capture audio.
   • The recorded audio will be saved and processed.

2. Transcription and Translation:

   • The app will identify the language, transcribe the speech, and translate it as needed.

3. Language Model Interaction:

   • The transcribed text will be used to generate a response using the integrated language model.
   • The response will be translated back to the detected language and provided as text and audio.

Contributing

Contributions are welcome! Please fork the repository and submit pull requests for any enhancements or bug fixes.

License

This project is licensed under the MIT License. See the LICENSE file for more details.