You need to agree to share your Hugging Face information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Access to this model is by request and is governed by the Hebrish terms of use (https://hebrish.app/terms). Non-commercial use is free; commercial use needs a commercial licence from access-requests@hebrish.app. Please also complete the attestation form: https://hebrish.app/legal/hebrish-terms-of-use.pdf

Jachnun v1.0 (ONNX): English-Hebrew code-switching

ONNX export of Hebrish/Jachnun, a fine-tune of NVIDIA Parakeet TDT 0.6B v2 for English speech with romanized Hebrew words dropped in ("pensia", "mechina", "makolet"). Read the main model card for what the fine-tune does, its results and its limits: it is a single-speaker proof of concept.

Website: hebrish.app: get the Hebrish keyboard for Android and the desktop apps, see the results, and find every model, dataset and demo in one place. Android updates through F-Droid: add the Hebrish F-Droid repository (https://fdroid.hebrish.app/repo; steps).

Access and terms of use. Use of this model and every converted or quantised build of it (GGUF, ONNX, sherpa-onnx, MLX) is governed by the Hebrish terms of use. Access is by request: use the button on this page, or write to access-requests@hebrish.app. Non-commercial use (personal, research, education, open-source projects) is free; please complete the attestation form. Commercial use, including telephony and call centres, needs a commercial licence: access-requests@hebrish.app. Versions released under CC-BY-4.0 (PoC-1 to PoC-5, v1.0) stay under that licence; the terms do not restrict rights it grants. The in-browser demo at hebrish.app/demo downloads these same files from Hebrish's CDN, cdn.hebrish.app.

The layout is the one used by onnx-asr and istupakov/parakeet-tdt-0.6b-v2-onnx, exported the same way (NeMo model.export) from training run r6-en30 (model Jachnun v1.0, 2026-10-19):

File	Precision
`encoder-model.onnx` + `encoder-model.onnx.data` (2.4 GB)	fp32
`decoder_joint-model.onnx`	fp32
`encoder-model.int8.onnx` (652 MB)	int8, dynamic (QUInt8)
`decoder_joint-model.int8.onnx` (9 MB)	int8, dynamic (QInt8)
`vocab.txt`, `config.json`	`nemo-conformer-tdt`, 128 mel features

Measured on a Ryzen 7 5700U CPU (2026-10-16): int8 transcribes a 5–10 s clip in 0.5–1.2 s.

Results

On the public benchmark, Jachnun v1.0 writes 73.3% of Hebrew terms exactly, against 32.6% for stock Parakeet v2 (scored with the main model, not this conversion). One speaker's voice: all training and test audio is the same speaker, so gains on other voices are untested. Full results: hebrish.app/benchmarks.

onnx-asr

# pip install onnx-asr[cpu] huggingface_hub
from huggingface_hub import snapshot_download
import onnx_asr

path = snapshot_download("Hebrish/Jachnun-ONNX",
                         allow_patterns=["*.int8.onnx", "vocab.txt", "config.json"])
model = onnx_asr.load_model("nemo-conformer-tdt", path, quantization="int8")
print(model.recognize("audio.wav"))  # 16 kHz mono

Drop allow_patterns and quantization for fp32.

Voxtype (push-to-talk dictation on Linux)

Voxtype loads this layout through its ONNX builds (sudo voxtype setup onnx --enable). It picks files by name, so give it a directory that holds one precision only:

hf download Hebrish/Jachnun-ONNX \
  --include "*.int8.onnx" vocab.txt config.json \
  --local-dir ~/.local/share/voxtype/models/parakeet-tdt-0.6b-v2-en-he-cs-int8

engine = "parakeet"

[parakeet]
model = "/home/<you>/.local/share/voxtype/models/parakeet-tdt-0.6b-v2-en-he-cs-int8"
model_type = "tdt"   # auto-detection looks for encoder-model.onnx, which the int8 directory lacks

Verified with Voxtype 1.1.1 on KDE Plasma Wayland (2026-10-16): model load 3.1 s, 0.53 s per clip.

Exact agreement with the PyTorch model has not been measured for this export; spot checks on four held-out clips came out right on the code-switched terms.

Other formats: GGUF (transcribe.cpp / Handy / Android) · sherpa-onnx int8 · NeMo original.

Licence: CC-BY-4.0, as for the base model (NVIDIA Parakeet TDT 0.6B v2).