Fact · Dictation software
Superwhisper's documentation says that a dictation can skip its language model stage, and that its Voice to Text mode outputs raw transcribed text with no language model involved.
Checked against its source on October 16, 2026
In the source's own words
“you can use the Voice to Text mode, which outputs raw transcribed text with no language model involved”
Sensitive Data Best Practices (Superwhisper documentation), Superwhisper, Superwhisper (vendor documentation), 2026, Section 'Transcription only (no AI post-processing)'.
Open the original- What it measures
- Superwhisper describes each dictation as two stages: a voice model that turns audio into text, and an optional language model that refines or reformats the text according to the mode. Voice to Text is the mode that skips the second stage.
- What it does not show
- This describes the app's configuration as the vendor documents it. It does not say how the modes differ in the words they keep, and the mode names and options may change between versions.
Source
Cited on
- AI Dictation: How AI Voice Typing WorksGuide
- AI Cleanup for Dictation: Fillers, Punctuation, RewritesGuide
- Do AI dictation apps send your words to a language model?Question
- How do you dictate with AI?Question
- Is there a dictation app without AI?Question
- Is transcription software considered generative AI?Question
- What is the difference between Wispr Flow and Superwhisper?Question
- Language model (in speech recognition)Glossary