V-Chatter is an AI-driven desktop pet application where users can upload custom character models (VRM files only currently) and interact with them in various ways!
Current features
Text and voice chat with user customized AI (UI will be improved in future)
"Vtuber" mode, which is a compact, resizable display window of the model
Music and video detection
Importing/replacing character model
Other features
- State animations
- Dance and pose selector
- "Mimic" feature where the user can upload audio and select the corresponding sentiment for the character model to say expressively
More about the AI chat feature
When the user submits a text or voice message, they will get an audio response if enabled. The character model will also pose and have a facial expression that corresponds to the underlying sentiment of the response.
There are two options for the AI feature. You can either use the built-in AI, or a python file I created that is publicly available to download from the Github repo attached to the manual link. I made the python version for the AI so users can further customize the personality of the AI, extend the AI's functionality, integrate custom local AI models, and choose the services that serve the AI functionality to reduce costs and/or improve quality. Default setup is OpenRouter for the large language models, OpenAI (built-in AI) or AssemblyAI (python version) for speech to text, and ElevenLabs for realistic & customizable text to speech. Additional info and support can be found in the official Discord server
Future Plans
I have plans to implement a lot more features that will improve everyone's experience with V-Chatter. If you wish to keep up to date with the latest developments or give suggestions of features you would like to see in the app, feel free to join the Discord Server
