Working on a accessible creative production suite featuring a voice-first multi-agent assistant. All core tools are completely free for hands-on use, while AI-powered automated generation runs on a flexible credit system with no subs.

So what started out as a text based chatbot project 8 months ago as my first ever project as a self taught coder is developing into something different. I've created an agent within my chat bot to help users create a product, using ElevenLabs V3 or OpenAI Realtime voice that works on a conversational basis rather than hardcoded commands

The agent can talk to you whilst your in chat or on a panel and navigate you to a particular panel if needed and throughout your session can select and substitutes models based on objectives such as quality or cost, proposes creative next steps, requests consent before paid inference, invokes generation, manipulates an editable multitrack timeline, and controls playback/time line like play video, delete my first image etc - through natural conversation.

Then if you wanted to create an image in another panel you can ask the agent via text or voice and they will navigate you to that panel and offer assistance their. Write your prompt for you and then even take that photo to the video suite to animate all using conversational language.

What do you think to this concept? I'm looking to further develop the idea across the platform to streamline some of the processes within it as my video demonstrates

This is my project i've been working on

Everything is a working concept and i'm just finalizing bits before release this week

• IDE Multi FIle Editor with AI assistant and live preview Split Screen Live Coding
• Multi Media Studio Editor
• Single Prompt to Full 2D and 3D Game Development Engine and Web Application Builder
• Video Editor with timeline controls, video effects, overlays, title, audio, podcast and music composer
• Music Studio with AI/Custom Lyrics
• Custom workspace environments with themes, live wallpapers, ambiant background tracks (Default options with light mode/dark mode with no wallpapers or music)
• Native 25+ Languages with RTL support. Already Hardcoded. Not live translated via web
• plus many more tools such as Podcast Creator with chat based/ custom context with 50+ voices and MP3 export.
• Full workflow tools like frame extract, analysis, transcribe, effects, file conversion audio analysis etc
• ...and of course the original chat bot interface that has cross device persistent multi model memory with vector base knowledge base via OpenAI and platform Drive storage.

You can start a conversation with any model on your laptop and next day carry on in a new conversation with another model on your phone with memory preserved across so you dont need to repeat yourself. The memory layer sits above the models entirely so is accessible by any LLM the platform supprts

Every tool, every feature i built will be completely free including GPT Nano, Gemini Flash and Deepseek.

Users can upload their own work to use for free and chat with selected free tier models with no limits.

If the user wants to generate a video or analyze a image, then that would be credit based. No subscription required and no tool access priorities over a non paying user.

Thats my concept i'm hoping to have launched in a few days and welcome any feedback/criticism you may have before i do launch.