Chat Flow: Support multimodal interactions #166

Summary

Enhance the existing chat flow to support multimodal interactions by enabling:

• Speech-to-Text (STT)
• Text-to-Speech (TTS)
• File-based inputs (Image, PDF, DOCs, etc.) through LLM calls

This will allow users to interact with the chat system using voice, documents, and media inputs instead of only plain text.

Scope

The overall implementation will include:

• Voice input support using STT
• Voice response support using TTS
• File upload and processing support for images, PDFs, DOCs, etc.
• LLM integration updates for multimodal processing
• UI and backend flow updates for handling these new interaction types