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: Sabich (recommended in the apps: code-switching plus tech vocabulary) and Jachnun (code-switching), Parakeet TDT 0.6B v2 fine-tunes (with GGUF, ONNX, sherpa-onnx and MLX builds); Shakshuka, an experimental Nemotron streaming fine-tune; Whisper-Hebrish; and a token classifier for romanized Hebrew.
• 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