#2626th of 38 in VoiceFor voice, speech and music

moonshine

On-device speech-to-text and voice agent toolkit with streaming models

Stars
11.3k
License
custom license
Last commit
Aug 2026
Last release
Aug 2026
Language
C++
Stars, last 29 weeks+3,812
31 Mar 2026: 7,450 stars7 Apr 2026: 7,532 stars14 Apr 2026: 7,619 stars21 Apr 2026: 7,728 stars28 Apr 2026: 7,791 stars5 May 2026: 7,852 stars12 May 2026: 7,945 stars19 May 2026: 8,077 stars26 May 2026: 8,182 stars2 Jun 2026: 8,259 stars9 Jun 2026: 8,325 stars16 Jun 2026: 8,403 stars23 Jun 2026: 8,456 stars30 Jun 2026: 8,501 stars7 Jul 2026: 8,563 stars14 Jul 2026: 8,625 stars21 Jul 2026: 10,176 stars28 Jul 2026: 10,442 stars4 Aug 2026: 10,547 stars11 Aug 2026: 10,735 stars18 Aug 2026: 10,842 stars25 Aug 2026: 10,894 stars1 Sept 2026: 10,948 stars8 Sept 2026: 11,005 stars15 Sept 2026: 11,056 stars22 Sept 2026: 11,109 stars29 Sept 2026: 11,142 stars6 Oct 2026: 11,177 stars13 Oct 2026: 11,218 stars20 Oct 2026: 11,262 stars

7,450 on 31 Mar 202611,262 on 20 Oct 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

Moonshine Voice is a C++ library with bindings for Python, JavaScript/WASM, iOS, Android, macOS, Linux, Windows and Raspberry Pi. It runs streaming speech-to-text models trained by the project, from tiny 1MB models up to ones the README says beat Whisper Large V3 on the Open ASR leaderboard. It also covers text to speech and conversational agents, all on-device without accounts or API keys.

Who it is for: Developers adding local real-time speech recognition or voice agents to apps

Strengths

  • Runs fully on-device, no account or API keys required
  • One library across Python, WASM, mobile, desktop and Raspberry Pi
  • Streaming design does work while the user is still speaking
  • Model sizes range from about 1MB up to Whisper Large V3-class accuracy

Weaknesses

  • Legacy non-English non-streaming models use a non-commercial community license
  • Repo license detected as NOASSERTION although README states MIT
  • RAM and latency figures are not stated in the README
  • No Docker image or live demo link in the README

What it needs

  • no GPU
  • Models: Moonshine speech-to-text models, Moonshine tiny 1MB models

Also in Voice

See all 38
Also in Voice
RankProjectScore
1voicestudioDesktop app for voice cloning, dubbing and dictation with local speech models79.5k starsAGPL-3.083 out of 100
2whisper.cppC/C++ port of OpenAI Whisper for local speech-to-text transcription54.5k starsMIT77 out of 100
3VoiceboxLocal voice studio for cloning, TTS, dictation and agent speech58.1k starsMIT71 out of 100
#4buzzOffline desktop app for transcribing and translating audio with Whisper22.1k starsMIT69 out of 100
#5meetilyDesktop app that records, transcribes and summarizes meetings locally Live demo ↗ (opens in a new tab)32k starsMIT68 out of 100