VoiceAssist-RAG is a multimodal voice-enabled assistant that combines speech recognition, retrieval-augmented generation (RAG), and text-to-speech to deliver intelligent, voice-based answers from domain-specific documents.
- 🔍 RAG Pipeline: Retrieval-Augmented Generation that provides grounded answers using
telecom_faq.txt - 🎤 Speech-to-Text (STT): Whisper-based transcription of voice input (
input.m4a,input.wav) - 🧠 Vector Search: Fast and accurate semantic search using FAISS
- 💬 Streamlit UI: Clean and interactive frontend for voice-to-answer interactions
- 🗣️ Text-to-Speech (TTS): Converts model-generated answers to audio output (
output.mp3) - 🧩 Modular Architecture: Easy to maintain and extend
- Python, Streamlit
- OpenAI Whisper for STT
- FAISS for vector-based semantic retrieval
- LLM (OpenAI / HuggingFace) for answer generation
- gTTS / pyttsx3 or similar for TTS
git clone https://github.com/yourusername/call-center-rag.git cd call-center-rag pip install -r requirements.txt
https://drive.google.com/file/d/1ePg8iGRVX87NQ8dEGb70B_8Fm4PGkJf6/view?usp=sharing