Ingutė
A second Lithuanian Piper voice — Reginutė's sister.
Lithuanian. Local. Free.
Full documentation in English: README_EN.md
A second Lithuanian Piper voice — Reginutė's sister.
Lithuanian. Local. Free.
Full documentation in English: README_EN.md
This is how she sounds. Synthesis, not a recording.
“Hi! I am Ingutė, Reginutė's sister. We are both digital Lithuanian voices for your projects that need a voice assistant. My sister and I are a little different, so choose which of us to use. We are both ready to work honestly on all your ideas.” (what she says, translated)
Her older sister Reginutė — hear her voice on her page.
This is a synthetic voice. All audio samples are generated by AI; they are not recordings of a person. If you publish content voiced with it, label it as machine-generated — the voice licence requires that, and the EU AI Act (Article 50, applicable since 2 August 2026) may require an AI label as well.
The voice lives on your computer or home server. No account, no API key, no monthly fee, no character limit. Your text goes nowhere.
Lithuanian has three pitch accents, and standard espeak-ng merges them into one. Ingutė, like Reginutė, takes the stress from a dictionary, not from espeak's guess.
„Susitinkam 15:00“, „nuo 2000 eurų“, „25 °C“ — numbers, times, units and abbreviations are expanded into words with the right Lithuanian endings before synthesis, not read digit by digit.
Ingutė shares the phonemizer, the dictionary and the synthesis wrapper with Reginutė, but speaks in her own voice. The same code serves both — only the model and its config change.
Since 20 October 2026 Ingutė is in the official Piper catalogue. She should appear in the Home Assistant Piper add-on once Piper 1.9.0 is out with the Lithuanian phonemizer. Until then — through the Wyoming server from the repository, about 5 minutes (README_EN, “Home Assistant”).
Free of charge; the licence does not forbid commercial use either. When you share the voice, pass on LICENSE-VOICE and state its origin. Read the whole licence before putting it into your product.
A voice assistant has two halves: a mouth and ears. Since 24 October 2026 there are ears that are fast on an ordinary CPU: LT_ausys. It is Kristijonas Jakubsonas's model kmynas (NVIDIA Parakeet, trained on LIEPA-3), converted by us to ONNX and compressed so that it runs without a graphics card. Model on Hugging Face, licence CC BY 4.0. Together with this voice it makes a complete Lithuanian voice assistant for Home Assistant that runs entirely at home.
Recognises a voice command in 0.21 s. Whisper-family models (Paprika included) take 3.55 s on the same mini PC: they encode a 30-second window for every phrase, however short.
4 CPU cores and 4 GB of RAM are enough — a mini PC or a Proxmox container. The model is 0.9 GB.
On read speech (FLEURS) it gets 10.6 % of words wrong, Paprika 12.0 %. Home Assistant's default Whisper base in Lithuanian: 88 %.
Everything runs on your own machine; nothing is sent anywhere. Home Assistant connects the ears the same Wyoming way as the voice.
Paprika is a Whisper model: it runs on faster-whisper, the same engine as the usual Home Assistant Whisper add-on. LT_ausys is a different kind of model (Parakeet, ONNX), and that add-on does not offer it by itself — our proposal to add it is still waiting for review. So the ears run in their own small Wyoming server (Python environment or Docker, about 5 minutes, recipe) and are added to Home Assistant as a separate Wyoming service. Paprika can stay next to it as a backup.
Our speaker, Alisa, has been listening with them since 24 October 2026. First day: 13 commands out of 13, each in about 0.3 s, without any extra training on the owner's voice. One home's experience, not a measurement.
More on Lithuanian speech recognition — setting it up in Home Assistant, the numbers and common questions — on the LT_ausys page.
Ingutė grew out of the Reginutė model and was
made with our own Python tools. Credit her like this:
Ingutė (lt_LT-ingute-medium), Robertas & Claude, grown from Reginutė.
Voice licence — NewGenLTU OpenRAIL-D 1.0 (LICENSE-VOICE); at its end is a list of ways the voice must not be used (for example, pretending to be a real person). Code — GPL-3.0.
You need the released piper-tts wheel — no fork, no patches. The voice also needs
the stress-aware phonemizer from the same folder: without it Piper would feed the model bare
letters and it would not speak Lithuanian.
Which files you need and a short Python example —
in the README; for Home Assistant — a
tested recipe in “Home Assistant”.
Model (.onnx, 63 MB) —
Hugging Face or
GitHub Release.
Not a programmer? You don't need to understand a single command. Paste this into any AI chat (Claude, ChatGPT or another) and keep asking in your own words:
Read https://raw.githubusercontent.com/RobertasTa/ingute/main/AI_CONSULTANT_BRIEF.md
and help me try the Lithuanian voice Ingutė.
That file lists what the assistant must ask before advising and what it must not claim — so that the advice rests on facts, not guesses.
Reginutė, and through her the LIEPA corpus of Vilnius University: Ingutė started from her.
Arūnas Smaliukas — for the stress dictionary that nearly every stressed word of the sisters stands on. Michael Hansen and the Open Home Foundation — for Piper.
No institution sponsors this project. Full credits with licences — at the end of the README.