Transcribro is an Android voice-input keyboard that converts speech to text entirely on the device. It runs OpenAI’s Whisper models through whisper.cpp, with Silero voice detection, so dictation needs no network connection. It can also act as the system speech-to-text service that other apps call.
Transcribro
github.com/soupslurpr/TranscribroA hardened pick. Worth the effort once you have chosen to shrink your footprint on purpose. Enough for most people. Threat levels
Voice typing is one of the quieter privacy holes on a phone, since the stock version ships your audio off to Google or Samsung to transcribe. Transcribro shuts that completely: the audio never leaves the device, and Whisper’s accuracy is good enough to rely on daily. The catches are real. It handles English only for now, and it is a young, single-developer project you install from Accrescent rather than the Play Store. If you dictate a lot and want it private, nothing else comes close.
Transcribro alternatives
A permissive license that is functionally identical to MIT, just with shorter wording.
Permits
- Commercial use
- Modification
- Distribution
- Private use
Requires
- License and copyright notice
Does not provide
- Liability cover
- Warranty
Why it matters: Permissive licensing lets anyone reuse this, including inside closed products. That is freedom to build on, but no guarantee that downstream copies stay open.
Plain-language summary of the project's license, not legal advice. Read the full text for the exact terms.
* Average ratings will show on this page once the threshold of 5 ratings is reached.