I wanted the chatbot to talk back.

A voice-enabled chatbot combining GPT-4 for conversation and DALL-E for image generation — an early, hands-on experiment in multimodal interaction.

GPT-4DALL-E

What it does

Voice in, conversation out — GPT-4 handling the exchange, DALL-E generating images on request inside the same interaction.

Architecture

The interesting part wasn't the model — it was everything wrapped around it: turning voice into a request, routing a conversation to text generation or image generation depending on what was asked, and returning a response in the right form.

Implementation

Built earlier, as hands-on exploration of multimodal AI interaction rather than a production product. The deeper implementation detail — request routing, session handling — isn't written up yet; worth expanding directly rather than guessing at here.

Trade-offs

Not documented. This one's here as a real, shipped experiment in stitching a model into a usable product, not because every design decision behind it was recorded at the time.