Do you want to see cloud Chat + TTS + STT models on OVS?

[debpalash]

Omnivoice was created as local first alternative for ElevenLabs even tho for lack of good translations locally i was prone to choose google/cloud translate.

Now as i'm getting few PR, issues for cloud Chat and TTS thoughts it's a high time we let community decide. We can make it as opt in/ module like translations so it only sends data if user explicitly asks for it.

What do you guys think?

[loryanstrant]

Personally, I don't have a need for translation functionality.
Additionally, I deployed this to replace my method of using voice clones with LocalAI + Qwen3-TTS, as OmniVoice is a better model.
(And my LocalAI+ Qwen3-TTS soution was to replace my ElevenLabs subscription!)
So, my preference is to stay local and have no cloud functionality.

[debpalash]

Thanks for sharing.

[xipb14]

ждем версию 0.3 в любом виде ))

[debpalash]

ждем версию 0.3 в любом виде ))

https://github.com/debpalash/OmniVoice-Studio/releases/tag/preview

[ratnerb88]

Hi everyone! I've been testing OVS for a long time and have a lot of thoughts to share. I believe LLM is absolutely necessary, including for translation and resegmentation. This is critical when using video dubbing. Furthermore, it needs extensive customization, similar to the pyvideotrans software. It turns out that I've been finishing OVS myself using vibecoding, but it still isn't working very well.
I still have a lot of thoughts and suggestions! Where can I share them directly with the developer?

[debpalash]

Hey, you can create an issue with feature request label with checklists of your ideas so we can track if they are addressed in future. You can also create a PR of your changes. Thanks for trying OVS

[PauloDand]

I totally see the need of cloud LLMs for people doing dubs and translations.
However, I feel like at this moment perhaps, OVS could use a bit of polishing on the voice management side first.
The voice cloning is incredibly good, but when you have several voices saved the UI gets totally out of hand.
A way to customize the saved voices with search, tags, favorites, and maybe a "profile pic" for changing voices at a glance would be, IMO, a feature that would help more people at this time.

[debpalash]

Thanks for input.
I'm planning a uiux redesign from a production usage standpoint. Soon will share on discord.