Offline voice (Parakeet)

Offline Parakeet is a dictation engine that runs entirely on your phone, with NVIDIA’s Parakeet TDT 0.6B v3, one of the most accurate open speech recognition models of its size. It transcribes 25 European languages and works out for itself which one you’re speaking, so a layout in any of them can dictate without choosing a model per language. Once the model is downloaded, no audio leaves the phone.

The languages: Bulgarian, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hungarian, Italian, Latvian, Lithuanian, Maltese, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swedish and Ukrainian. While a layout in another language is in use, dictation uses the system recognizer instead, the same way offline Whistle hands over its other languages.

Setting it up

WM Keyboard › Voice typing

Choose Offline Parakeet under Recognition engine, then download the model from the Parakeet model group that appears below it. It is large: 719 MB, the 4-bit build made for phones, and it suits a recent phone with plenty of memory. On mobile data the download asks first. It comes from Hugging Face at a fixed revision, and both of its files are checked against their published SHA-256 before they count.

How it works

Like Whisper, Parakeet transcribes a phrase once you stop, and a phrase can run to 50 seconds. Unlike Whisper it doesn’t write its answer a word at a time: an encoder reads the whole clip once, and a small transducer then walks through it, writing a word piece where it hears one and skipping ahead where the model’s own duration head says nothing more is coming. That is most of why it’s quick for its size. The first phrase after the keyboard starts takes a few seconds longer while the model’s weights are prepared.

The text arrives punctuated and capitalised by the model itself, and then goes through the same spacing, spoken punctuation and undo as any other dictation (see Voice typing).

Details & edge cases

Several languages at once. A dictation that listens for several languages (see Language and engine) is Parakeet’s when every one of them is among its 25; it tells them apart on its own.

No words to listen for. Parakeet has no way to take a list of names or terms, so Words to listen for does not show while it is the engine.

Both editions. The runtime, ExecuTorch built for the keyboard’s models alone, is in full and lite alike. A build made without it, as F-Droid’s is, does not offer the engine.

The licence. The model is NVIDIA’s, under CC-BY-4.0; it is credited in the app’s About screen.