Hebrish · עבריש

Dictation that gets the Hebrew words right.

Ordinary dictation hears makolet and writes "Micollette". Hebrish writes the Hebrew word.

Hebrish improves ASR accuracy for code-switched English–Hebrew, for English-speaking immigrants in Israel and users throughout the Jewish world. It is a one-person, not-for-profit project by Daniel Rosehill, and its models and data are open.

• Models: see them all on one page in the Hebrish Models Space.
  • Sabich, recommended in the apps: code-switching plus tech vocabulary. It comes in variants for every size of device: Qwen3-ASR-1.7B and Canary-Qwen-2.5B for desktop GPUs, Parakeet-110M for small devices, and asr4all-S and -M for wearables.
  • Jachnun: code-switching only.
  • Shakshuka: streaming, so the words appear while you speak.
  • Whisper Hebrish, and a token classifier for romanized Hebrew.
  • On-device builds (GGUF, ONNX, sherpa-onnx, MLX) are in -<format> repos. Which model should I use?
• Data: read-aloud code-switching sentences and a frozen benchmark, scored by Hebrew-term recall on a clean subset. Results: hebrish.app/benchmarks.
• Apps: Hebrish runs the models on-device on Android, Linux and Windows.

Everything in one place, including repos still on Daniel's personal account: the Hebrish collection.

Website: hebrish.app · GitHub: Hebrish-ASR