Speech recognition
Local speech recognition uses whisper.cpp / ggml / OpenAI Whisper — MIT.
Acknowledgments
Open-source licenses and credits.
Diktator is built with open-source speech technology and language data. These credits describe the components used in the current Mac, iPhone, and iPad editions.
Direct Mac updates
The direct-download Mac edition uses signed updates through Sparkle 2 — MIT.
iPhone and iPad language data
The mobile keyboard language packs include derived data from these projects:
- CC-CEDICT — Chinese dictionary — CC BY-SA 4.0.
- jieba — Chinese word frequencies — MIT.
- FrequencyWords — word frequencies — CC BY-SA 4.0.
- Leipzig Corpora Collection — next-word data — CC BY.
- Mozc — Japanese kanji data — IPAdic notice and public-domain portions as described upstream.
- OPUS · OpenSubtitles 2018 · OpenSubtitles.org — Japanese next-word frequencies derived from the Japanese corpus.
Pierre Lison and Jörg Tiedemann (2016), OpenSubtitles2016: Extracting Large Parallel Corpora from Movie and TV Subtitles. Proceedings of the 10th International Conference on Language Resources and Evaluation (LREC 2016).
- Unicode CLDR — emoji annotation reference data — Unicode License v3.
About these notices
Project and license names remain in their official form. Links lead to the upstream project or license notice. Apple frameworks and services are governed by Apple's platform terms and are not presented here as third-party open-source components.