#4343rd of 93 in VoiceFor voice, speech and music

openless

Push-to-talk dictation that inserts AI-polished text at the cursor

Stars
4.1k
License
AGPL-3.0
Last commit
Nov 2026
Last release
Nov 2026
Language
Rust
Stars, last 28 weeks+3,692
2 May 2026: 437 stars9 May 2026: 1,122 stars16 May 2026: 1,345 stars23 May 2026: 1,567 stars30 May 2026: 1,859 stars6 Jun 2026: 2,151 stars13 Jun 2026: 2,278 stars20 Jun 2026: 2,404 stars27 Jun 2026: 2,488 stars4 Jul 2026: 2,589 stars11 Jul 2026: 2,697 stars18 Jul 2026: 2,764 stars25 Jul 2026: 2,829 stars1 Aug 2026: 2,895 stars8 Aug 2026: 2,968 stars15 Aug 2026: 3,013 stars22 Aug 2026: 3,191 stars29 Aug 2026: 3,348 stars5 Sept 2026: 3,414 stars12 Sept 2026: 3,495 stars19 Sept 2026: 3,587 stars26 Sept 2026: 3,645 stars3 Oct 2026: 3,712 stars10 Oct 2026: 3,781 stars17 Oct 2026: 3,885 stars24 Oct 2026: 3,969 stars31 Oct 2026: 4,032 stars7 Nov 2026: 4,082 stars14 Nov 2026: 4,129 stars

437 on 2 May 20264,129 on 14 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

OpenLess is a Tauri 2 and Rust voice-input app for macOS, Windows and Linux, with Android and iOS apps also in the repo. You hold a global hotkey and speak. It transcribes, rewrites the text in a chosen style, and types it at the cursor, falling back to the clipboard if insertion fails. ASR and polish models are bring-your-own: cloud providers, any OpenAI-compatible endpoint, or local Qwen3-ASR on macOS.

Who it is for: People who dictate prompts and emails into any app using their own models

Its own README calls it an alternative to Wispr Flow.

Ranked among open-source alternatives to Wispr Flow (#3 of 3).

Strengths

  • Many ASR and polish providers plus any OpenAI-compatible endpoint; no single vendor required
  • Four output modes, including a structured AI-prompt mode, and custom style packs switchable by hotkey
  • Streaming insertion to the cursor, with automatic one-shot paste fallback
  • Local ASR option: bundled Qwen3-ASR 0.6B/1.7B on macOS, Foundry Local Whisper on Windows

Weaknesses

  • Linux uses a separate native host and has no in-app updater
  • macOS build is ad-hoc signed; needs xattr -cr to bypass Gatekeeper
  • Project is in a stability freeze and rejects new-feature PRs
  • Windows local ASR is experimental; most providers are cloud services

What it needs

  • Needs Cloud ASR/LLM API credentials (unless using local ASR), Microphone and accessibility permissions
  • Models: Qwen3-ASR 0.6B/1.7B, Whisper (Foundry Local), sherpa-onnx, OpenAI-compatible endpoints
  • README: alternative to Wispr Flow

What changed

From the release notes. No flag means the note does not say so, not that the upgrade is safe.

Also in Voice

See all 93
Also in Voice
RankProjectScore
1voicestudioDesktop app for voice cloning, dubbing, dictation and audiobooks113k starsAGPL-3.098 out of 100
2handyOffline push-to-talk dictation app that pastes transcripts into any text field39.1k starsMIT92 out of 100
3whisper.cppC/C++ speech-to-text engine running OpenAI Whisper models locally55.3k starsMIT90 out of 100
#4meetilyDesktop app that records, transcribes and summarizes meetings locally Live demo ↗ (opens in a new tab)32.8k starsMIT87 out of 100
#5buzzOffline desktop app for transcribing and translating audio with Whisper22.9k starsMIT84 out of 100