Back to browse
I built an open source dication tool based on benchmarks

I built an open source dication tool based on benchmarks

by emillykkeg·May 14, 2026·1 point·0 comments

AI Analysis

●●SolidSolve My ProblemDark Horse

Benchmarks 34 local speech models so you don't have to guess which one works.

Strengths
  • Actual WER and speed benchmarks for 34 on-device models, not just marketing claims.
  • Runs fully offline with no account or cloud dependency, unlike Whisper Desktop clones.
  • Adapts to custom vocabulary and fixes near-misses automatically based on user feedback.
Weaknesses
  • No Linux support yet, and punctuation accuracy isn't included in benchmark scores.
  • Another local Whisper GUI in a space with Whisper Desktop, Buzz, and MacWhisper already.
Category
Target Audience

Developers and privacy-conscious users needing offline dictation

Similar To

Whisper Desktop · Buzz · MacWhisper

Post Description

I've built an AI dictation tool (like many others), but this time I have taken the time to benchmark all of the 34 models that we provide, so users can actually make a qualified choice on what model should be the daily driver.

The best is that it's open source, so if you think we can improve something, please come and help out.

Similar Projects

AI/ML●●Solid

MoonshineFlow

Local Moonshine transcription streams text to any app, no audio leaves your Mac.

CozySolve My Problem
jrm-veris
602mo ago