-
TODO Create menu images, add to menu_images folder under data for the frontend
-
TODO Could add a 'Show me what this drink looks like' feature, which upon user request, the application would pop up an image of the menu item.
-
TODO Build out the Azure Backend Toggle to switch from the Realtime API to the STT-LLM-TTS services pipeline to compare the latency, phonetics, conversable AI experiences
-
TODO Deploy it
-
TODO Add in tokens for session; research audio tokenizer from openai if it is openly available
-
TODO Add in Azure AI Speech Backend
-
TODO Update Readme to include Azure AI Speech backend separate process
-
TODO Update Readme with latest Architecture Diagram that covers Azure AI Speech backend process
- TODO Could add in Bing API for searching latest coffee info on the web
- TODO Create audio actions/tools/functions to open and close chat history as well as show menu
- TODO Consider adding a chat box to the bottom of Transcript History. This would make it a text chat experience, instead of audio Transcript History