Fact · Dictation software
TypeLeast's home page states 0 bytes of recordings are uploaded, and that recordings stay in memory and are discarded after recognition.
Quote found word for word on the live source, last checked October 25, 2026
In the source's own words
“0 字节 上传的录音。 语音在你的电脑上识别。发给模型的，只有文字。 录音不落地。 只在内存里，识别完就丢弃。”
TypeLeast: home page (说出来，就写好了), TypeLeast, 2026, Privacy section of the home page.
Open the original- What it measures
- The privacy section gives the three guarantees in order: no uploaded recordings, speech recognized on the computer, and text-only requests to the model.
- What it does not show
- The page makes the claim about uploads and memory. It does not state a retention period for the text history, which it says is encrypted and can be turned off.
In the data
A row of Dictation App Features Compared: Offline, Privacy, AI, as TypeLeast, 0 bytes uploaded: 0 1 = yes, 0 = no.
- App
- TypeLeast
- Feature
- Keeps audio on the maker's servers
- Detail
- 0 bytes of recordings uploaded
- As of
- 2026-10
- Note
- 0 means the maker's page says it does not keep audio on its servers.