voice_audio_multi/tools/community_models/convert_underdog_bark3_standard.py
Fabio c36510c21e
Some checks failed
Linux Build / Linux cpu (push) Has been cancelled
Linux Build / Linux cpu-clang (push) Has been cancelled
Linux Build / Linux vulkan (push) Has been cancelled
mac-build / macOS CPU (push) Has been cancelled
Nix Build / Nix (cpu) (push) Has been cancelled
Nix Build / Nix (vulkan) (push) Has been cancelled
windows-build / Windows CPU (push) Has been cancelled
Initial import: audio.cpp with Orange Pi documentation
2026-11-04 19:05:22 +01:00

20 lines
956 B
Python

#!/usr/bin/env python3
"""Convert the pinned standard Underdog Bark 3.0 checkpoint to native GGUF.
Reuses the packed Bark reader and precision policy; the immutable profile keeps
standard/Plus checkpoints, variants and provenance separate. Python/ANS is only
needed for conversion, never native inference.
"""
from convert_underdog_bark3_plus import BarkPackedProfile, main
SOURCE_REPO = "ConwayResearch/Underdog-Bark-3.0"
SOURCE_REVISION = "85db2ad18ab41e534a222ee253c29191d77ed919"
SOURCE_SHA256 = "c18de9418d73cbeb80b8c1287b392d480a56e191ef4a81d1baaa1b02e66d92ed"
VARIANT = "underdog_bark3_standard"
STANDARD_PROFILE = BarkPackedProfile(SOURCE_REPO, SOURCE_REVISION, SOURCE_SHA256,
VARIANT, "Underdog Bark 3.0", "docs/models/underdog_bark3.md",
bf16_f32_prefixes=("decoder.", "encoder_projector.", "joint.head."))
if __name__ == "__main__":
main(STANDARD_PROFILE, __doc__)