Yanis Lukes comments

Results 59 comments of


                                            Yanis Lukes

Merging "Equipped X, Unequipped X, Equipped Y, Unequipped Y"

Also currently only 5 events are sent to LLM, it should only be done for the 5 events after merge cleanup. Otherwise more important events would be lost.

Switch voice model while waiting on LLM's response

I had tried to do this by just sending the [voice model change] request to xVASynth without waiting for a response. Though often the LLM response would be ready before...

Switch voice model while waiting on LLM's response

STT? This is about TTS/T2S. The needed TTS voice model is unknown until an NPC is chosen.

Add Push to Talk

Found a commit cuneyttyler, but it awaits for Mantella Spell to end listening. https://github.com/cuneyttyler/Mantella/commit/78c1eaf64f3b256c40b3884a8eb8e74bee656bda

Update client.py: raise error on 429 get_config

Linter test failure: I simply copied the code from the same file, which also has the exact same linter issue: 😤️ https://github.com/gradio-app/gradio/blob/e0f24d6c3513fdc595d33bfd38d2f42bdb90d820/client/python/gradio_client/client.py#L1262-L1266

HF Gradio demo: sudden gender flip for slider

Ok, so I am not going crazy. Also cloning never works for me. It still seems to take the generated artifical speaker. I am thinking of using multiple voices and...

MeloTTS?

Added to converted HF dataset, anyone can make the PR if they want to: https://huggingface.co/datasets/Pendrokar/open_tts_tracker

Add Chatterbox

Good job! Added to the cloned HF repo. https://huggingface.co/datasets/Pendrokar/open_tts_tracker

Incompatible with 50XX RTX series GPUs

Yes, but you will need to keep running xVASynth from a development environment. "Yeah it’s the PyTorch that you pip install —pre torch torchaudio index-url to the nightly. Although on...