Audio and Image processing inside conversations

Hello everyone;

I'm not sure if this feature has been requested before, but I'm writing it here anyway.

When a client sends an audio message, an image, or even a document, its contents are unknown unless a human reviews the conversation. The AI ​​assistant is currently blind to these messages, so... Is it possible to include a feature to send this content to an AI engine for processing, or to process it directly within Chatwoot?

It would be fantastic if the assistant could "see" all types of messages.

Best regards and much success to all.