Cloud & Local Voice Assistant with RAG and TTS
Jan 2026
Finished
Full-stack voice assistant pipeline: Speech-to-Text (Whisper) → Text Processing → Information Retrieval (local knowledge base or Tavily web RAG) → Text-to-Speech (Qwen3-TTS 1.7B with custom voice selection). The TTS engine supports 10 languages (Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, Italian) with 9 different voice personalities. Runs both as a CLI tool and as an interactive Streamlit web application. Requires SoX for audio processing. Modular architecture with editable pip install.
Technologies
AI
HuggingFace
Jupyter
LangChain
NumPy
Python
PyTorch
RAG
