Model · speech-to-text
Boxty v1.0: Parakeet TDT 0.6B v2 for English and tech dictation
English dictation without Hebrew, in my voice: model and product names, CLI and cloud terms, brands, acronyms, and numbers and currency in digits.
Which model? Boxty has none of the code-switching vocabulary. Use Jachnun if you mix Hebrew words into English; use Boxty for plain English and tech dictation. See Jachnun →
About the model
Boxty is a fine-tune of NVIDIA Parakeet TDT 0.6B v2 for English dictation with no Hebrew in it. It targets tech vocabulary (model and product names, CLI and cloud terms), brand names, acronyms, numbers and currency written in digits, homophones, Irish-English phrasing, and dictated emails and messages, all in my voice.
It is the sister model to Jachnun, the English–Hebrew code-switching fine-tune from the same base. Boxty has none of the code-switching vocabulary, so code-switched speech stays at stock level on purpose. If you mix Hebrew into your English, use Jachnun.
Boxty is named after the Irish potato pancake. The no-Hebrew line uses Irish food names; the code-switching line uses Israeli ones.
You can use it through Parlance, my general-purpose on-device dictation app for Android and Linux, whose model list includes Boxty.
Which file should I download?
The model repo is gated: request access on Hugging Face. The links below work once access is granted and you are signed in.
If unsure, take Q8_0. On desktop CPUs it was also faster than Q4_K_M (about 3× realtime against 1.4× on 8 cores), and both gave the same transcripts as the full model on the test clips.
GGUF Q8_0 730 MB
Android phone or Linux desktop. Take this one if unsure. Needs Hugging Face access to the gated repo.
Run with Parlance / Offline Voice Input (Android) / Handy (Linux) Download GGUF Q8_0 →GGUF Q4_K_M 475 MB
Phones with 4 GB of RAM or less, or if Q8_0 is too slow.
Run with Offline Voice Input Download GGUF Q4_K_M →.nemo checkpoint 2.47 GB
Python, GPU inference or further fine-tuning, at full precision.
Run with NVIDIA NeMo Download .nemo checkpoint →On Android
- Install Offline Voice Input (source), download the Q8_0 GGUF on the phone, import it in the app and select it.
- Use it as the system voice input, or through a keyboard whose mic key opens it: SwiftKey, HeliBoard or FlorisBoard. Gboard's mic button cannot hand speech to it.
- Set the app's battery optimisation to Unrestricted so Android does not kill it between dictations.
On Linux
- Handy loads the same GGUF. Copy it into
~/.local/share/com.pais.handy/models/under its own file name.
Results
v1.0 · Boxty, run r7-general, evaluated 20 October 2026. Lower WER is better; a higher share of terms correct is better.
| Test set | Stock v2 | Fine-tune v1.0 · Boxty |
|---|---|---|
| Hebrew-free English (normalised WER) | 1.79% | 1.04% |
| Tech dictation (normalised WER) | 2.51% | 1.35% |
| LibriSpeech (normalised WER) | 1.23% | 1.18% |
| Code-switched speech (normalised WER) | 11.5% | stock level, on purpose |
- Against Jachnun on the same tests: Hebrew-free English 1.04% against 1.23%; tech dictation 1.35% against 1.74%.
- Code-switched speech stays at stock level (11.5%) because the training data has no Hebrew in it.
Training
- 142.5 min of my own recordings with no Hebrew in them, plus 15.9 min of LibriSpeech replay to keep general English.
- Fine-tuned from stock Parakeet TDT 0.6B v2, the same base as Jachnun.
Training data
Checkpoints and roadmap
Where the model is now and the checkpoints still to come. Planned stages are the current plan, not promises.
- v1.0 · Boxty Done 20 October 2026
First release: 142.5 min of Hebrew-free recordings plus 15.9 min of LibriSpeech replay.
Limitations
- One speaker: an English speaker with an Irish accent living in Jerusalem. Gains on other voices are untested.
- No code-switching vocabulary: Hebrew words in English speech come out at stock level. Use Jachnun for that.
- The model repo is gated, with access by request.
Related
- Boxty v1.0 write-up (blog)What Boxty is for and how it compares