You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Proposing a URML v0.1 capability-manifest mapping for OpenVoice over myshell-ai/OpenVoice. URML (Apache-2.0) is a substrate-neutral spec for robot intent: typed primitive vocabulary + capability manifest + static validator.
This is URML's first TTS RFC. URML's Layer-2 speak primitive renders text output to audio; the zero-shot voice-cloning angle is the OpenVoice contribution URML wants to declare — a single reference clip lets a robot fleet share one consistent voice identity across deployments, and the cross-lingual capability pairs directly with URML's Layer-4 multilingual structural-slot reservation.
This is proposal-only, part of URML's Move #12 outreach (16 RFCs covering speech / translation / robot-command-library substrates for URML's NL layer).
TTS-engine-class declaration shape. Does the OpenVoice team have a preferred convention for declaring "OpenVoice is the TTS engine" in a downstream manifest, or is this internal detail?
Voice-clone-reference declaration. Is a manifest field that names the reference voice clip URI useful (for downstream consent / audit), or does it introduce a privacy footprint the project would rather not have associated with it?
Voice-style enumeration. Is the preset set stable enough for URML's manifest to declare an enum, or is it evolving fast enough that a free-form string is the right shape?
Multilingual coverage. OpenVoice supports cross-lingual cloning. What is the canonical set of synthesis languages URML's manifest should list (README authoritative)?
Adapter home. URML-side adapter in URML's reference/speech-bridge/, contributed example in OpenVoice/examples/, or external bridge repo?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Hi @myshell-ai/OpenVoice team,
Proposing a URML v0.1 capability-manifest mapping for OpenVoice over
myshell-ai/OpenVoice. URML (Apache-2.0) is a substrate-neutral spec for robot intent: typed primitive vocabulary + capability manifest + static validator.This is URML's first TTS RFC. URML's Layer-2
speakprimitive renders text output to audio; the zero-shot voice-cloning angle is the OpenVoice contribution URML wants to declare — a single reference clip lets a robot fleet share one consistent voice identity across deployments, and the cross-lingual capability pairs directly with URML's Layer-4 multilingual structural-slot reservation.This is proposal-only, part of URML's Move #12 outreach (16 RFCs covering speech / translation / robot-command-library substrates for URML's NL layer).
Full RFC with manifest mapping, three alternatives, and the voice-clone-reference design discussion: https://github.com/URML-MARS/URML/blob/main/docs/rfcs/0156-openvoice-outreach.md
Questions worth maintainer input on:
reference/speech-bridge/, contributed example inOpenVoice/examples/, or external bridge repo?Ido Yahalomi (URML maintainer, urml.dev, greenvh@gmail.com)
AI-assisted prose, maintainer-reviewed before posting (see VIBE.md). Human-only correspondence available on request.
All reactions