YoozTextCleanup

Prebuilt Swift XCFramework for the Yooz text-cleanup engine (libyooz_text_cleanup), part of the Yooz ecosystem. It provides rule-based post-processing for on-device speech-to-text output (number, decimal, and version normalization and related cleanup), exposed to Swift via UniFFI.

This repository hosts the compiled binary artifact consumed by the yooz-engine SwiftPM package as a remote .binaryTarget. It is not model weights; it is native Rust compiled for Apple platforms.

Contents

File Description
YoozTextCleanup-0.6.6-07ff96f.xcframework.zip Universal (arm64 + x86_64) macOS XCFramework
  • Crate version: 0.6.6
  • Source provenance: yooz-labs/yooz-engine text-cleanup/, unchanged since commit 07ff96f
  • SHA-256: c859400999a7af479303316440032f6323dcfc56ad15a9295f92f9b3bec843a5
  • Platforms: macOS (arm64, x86_64)

Consuming from SwiftPM

.binaryTarget(
    name: "YoozTextCleanup",
    url: "https://huggingface.co/YoozLabs/YoozTextCleanup/resolve/main/YoozTextCleanup-0.6.6-07ff96f.xcframework.zip",
    checksum: "c859400999a7af479303316440032f6323dcfc56ad15a9295f92f9b3bec843a5"
)

The checksum pins content integrity; the versioned filename is never overwritten in place, so the main resolve URL is effectively immutable for a given version.

License

Compiled from source licensed under PolyForm Shield 1.0.0 (source-available). See the yooz-engine repository.

Contact

dev@yooz.info

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support