Textify’s desktop app uses Electron and Chromium, React, uiohook-napi and libuiohook, and the Delta Chat fork of dbus-next for its interface and desktop integration.
Local speech recognition uses ggml-org’s whisper.cpp, handy-computer’s transcribe.cpp, and ShugoAI’s audio.cpp. Whisper was developed by OpenAI, Parakeet by NVIDIA, Qwen3-ASR by the Qwen team, and Confucius4-R2T2 by NetEase Youdao. Textify also credits the publishers of the model conversions identified in its signed catalogs.
Model binaries are downloaded separately. Exact revisions, checksums, licenses, and provenance are recorded in the signed model catalogs. Availability varies by platform; see the desktop guide.
The desktop notices describe the software and license texts included in the current app. Root third-party notices and copied license texts retain attribution for the shared catalogs and historical model evaluation, including models and runtimes that are not exposed by the current desktop app.