Skip to content

docs(voice): update the voice typing section to the current menus, models and options - #830

Merged
LeanBitLab merged 2 commits into
mainfrom
docs/voice-features-section
Nov 15, 2026
Merged

LeanBitLab merged 2 commits into
mainfrom
docs/voice-features-section

Conversation

@LeanBitLab

Copy link
Copy Markdown
Owner

What

The voice section of FEATURES.md and the voice setup steps in the README were out of date:

  • menu names: now Plugins > Voice Input > Whisper Models > Custom GGML Model (as in the app)
  • models: the official LeanType models (tiny 31 MB, base 57 MB, small 182 MB, English and multilingual), custom import for others such as FUTO ACFT
  • options: silence timeout (default 3 s), transcribe while speaking, keep screen on, keep model in memory, the voice key while processing
  • new requirements and battery notes: Android 5.0+, CPU requirement of the 64-bit plugin build, on-demand binding

Notes

  • The CPU requirement comes from the plugin build flags (-march=armv8.2-a+dotprod+fp16 since plugin v1.0.3). It is not tested on an older device.
  • Docs only, so the build workflow does not run.

…dels and options

Use the menu names of the app (Plugins, Voice Input, Whisper Models, Custom GGML Model),
list the official LeanType models and their sizes (31, 57 and 182 MB, English and
multilingual), describe the options added in 4.3.x (silence timeout, transcribe while
speaking, keep screen on), and add the plugin requirements and the battery behavior.
@LeanBitLab
LeanBitLab merged commit 7fc9bab into main Nov 15, 2026
@LeanBitLab
LeanBitLab deleted the docs/voice-features-section branch November 15, 2026 12:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant

⌂ Proxy