renikud-plus 0.2.0

ONNX Runtime inference for Hebrew grapheme-to-phoneme conversion (ReNikud Plus)

ReNikud Plus — Hebrew Grapheme-to-Phoneme Inference

Convert unvocalized Hebrew text into IPA for TTS, speech technology, and spoken-language research.

Benchmark

Install

This repository contains only the code required for ONNX inference. It has no training pipeline, checkpoint files, or PyTorch dependency.

pip install renikud-plus
hf download notmax123/RenikudPlus model.onnx --local-dir .

Or from a clone of this repo:

uv sync
hf download notmax123/RenikudPlus model.onnx --local-dir .

Usage

from renikud_onnx import G2P

g2p = G2P("model.onnx")
print(g2p.phonemize("שלום עולם"))
# → ʃlˈom ʔolˈam

For a gender-conditioned ONNX model, pass speaker and target_speaker as 0 (unknown), 1 (male), or 2 (female):

g2p.phonemize("היא רצה", speaker=2, target_speaker=2)

Niqqud output

vocalize renders the same predictions as pointed Hebrew (niqqud) instead of IPA — for TTS engines that read niqqud natively but ignore phoneme markup. It accepts the same speaker / target_speaker arguments.

print(g2p.vocalize("שלום עולם"))
# → שׁלוֹם עוֹלַם

Niqqud has no stress mark, so predicted stress is not represented in this output (it is in phonemize). Diacritization is phonetically faithful but not publication-grade — e.g. shva in clusters is omitted.

Citation

@misc{melichov2026renikud,
  title={ReNikud: Audio-Supervised Hebrew Grapheme-to-Phoneme Conversion},
  author={Maxim Melichov and Yakov Kolani and Morris Alper},
  year={2026},
  url={https://arxiv.org/pdf/2606.20179},
}