itch.io

Are there plans on adding Kokoro or Kitten TTS as an alternative which would run locally instead of relying on some thirdparty app? (Kokoro and Kitten TTS are both very light weight models)

Voxta has been dealing with LLMs determining what actions to take etc. and other things related to a character you can interact with. You could probably join the Discord server and ask for advice there from Acidbubbles and others. https://discord.gg/voxta

Any plans to add LLM(AI) integration to this? Preferably via some API like OpenAI compatible one or OpenRouter/Koboldcpp one. (there are also some lightweight TTS available like Kokoro, and Whisper for STT)

Does the OpenAI API support mean you can use custom endpoints/proxies? That would allow for local LLMs to be used if that was the case.

Most LLMs use the same or similar architectures, so that shouldn't be a big problem. But you'd still have the option to pay if you don't know what you're doing. And using OpenAI compatible API would be the broadest way to support multiple GUIs for LLMs.

Sure if you aren't tech savvy you can always just pay for the ease of use but if you don't mind figuring things out to do it for free then you should still be allowed to do that.

(Btw you don't need particularly beefy computers to run LLMs especially llamacpp.. it can run on your phone if you have 6-8gb memory.)

OpenRouter should a pretty easy option since it also uses OpenAI's API. (also consider local options for LLMs like koboldcpp/text-generation-webui) Another thing is speech to text, you could use WhisperLive/fasterwhisper for local or other APIs like Deepgram, Azure.

You do realize they don't need to force you to use OpenAI? They could just open up the API so that you could use locally run LLMs.. which could keep the game alive long after the devs stop supporting the game.
OpenAI's API is pretty common nowadays and many popular local LLM UIs support it like text-generation-webui.. though Koboldcpp which is probably the easiest to use uses it's own API, but should be fairly easy to implement still. (the devs could easily look at things like SillyTavern and the code on how they implemented those things)

If you ever plan on supporting local models, you might want to use the OpenAI API with the ability to change the proxy server so it can be used by many different open source LLM backends like Oobabooga's text-generation-webui etc. (Koboldcpp had it's own API but Ollama has partial OpenAI API support)