Credits & licences
OpenRead stands on work by other people. This page lists what the app is built from, and the terms that come with it.
Preset voices (14)
The fourteen preset Vietnamese voices come from the VieNeu-TTS preset collection by Phạm Nguyễn Ngọc Bảo (pnnbao-ump), whose project is licensed Apache-2.0. Upstream attaches no separate non-commercial term to these presets. The credit stands regardless — the work is theirs.
VieNeu TTS
The Vietnamese speech model that turns text into audio on your device.
sea-g2p
Grapheme-to-phoneme conversion for Vietnamese, licensed Apache-2.0, ported to a Rust library inside the app.
MOSS Audio Tokenizer Nano
The neural audio codec that decodes generated tokens back into a waveform.
ONNX Runtime
Runs the models on-device, licensed MIT, with its own bundled third-party notices.
Supertonic 3 — multilingual voice pack
An optional download for reading languages other than Vietnamese, based on the Supertonic 3 model by Supertone Inc., licensed BigScience OpenRAIL-M. OpenRead stores its weights as int8 and repackages it. The licence carries use restrictions you must accept before downloading — see the voice pack terms. Its text normalisation rules are ported from Supertone’s MIT-licensed sample code. OpenRead is not affiliated with Supertone Inc.
Each component ships with the complete licence and notice files from the exact release used. If you believe anything here is credited incorrectly, please write to us.