Official documentation
FlashVoice README (English), GitHub
GitHub, harry0703/FlashVoice (maker's project repository), 2026
Read the originalRead on October 22, 2026
- What it is
- The English README of FlashVoice, a macOS and Windows dictation app from its own repository. It describes on-device SenseVoice recognition, typing into any app, custom vocabulary, optional AI proofreading and offline use after the model downloads.
- How much weight it carries
- The project's own README, read on the access date. It is the maker's documentation, not a product page, and it has no pricing; it does not claim anything about training on recordings.
Facts taken from this source
- FlashVoice's README says optional AI proofreading can add punctuation, polish wording and correct recognition errors.
“Optional AI proofreading can add punctuation, polish your wording, and correct recognition errors.” (Introduction)
- FlashVoice's README lists custom vocabulary among its practical tools.
“Use custom vocabulary, transcription history, and multilingual input” (Highlights, 'Practical tools')
- FlashVoice's README says it transcribes with SenseVoice without uploading recordings, and that audio and video files stay on the device.
“Transcribe with SenseVoice without uploading your recordings ... Your audio and video files stay on your device and are never uploaded for transcription.” (Sections 'Highlights' and 'Privacy and offline use')
- FlashVoice's README says you place the cursor in any text field, press a shortcut, and FlashVoice types as you speak.
“Place the cursor in any text field, press a global shortcut, and FlashVoice types as you speak.” (Introduction and 'Getting started')
- FlashVoice's README says its core features keep working without an internet connection once the model is downloaded.
“Once the model is downloaded, the core features continue to work without an internet connection.” (Section 'Privacy and offline use')