# moonshine-ai/moonshine

> Open source AI toolkit for building real-time voice agents and interfaces: on-device speech to text, intent recognition, and text to speech with low latency. Supports Python, JavaScript/WASM, iOS, Android, macOS, Linux, Windows, and Raspberry Pi.

- Magnitude: 2.8 out of 10 — Breakout
- Stars: 11,366 total · +23 stars measured 2026-11-01, 00:07–15:54 UTC, ≈ 33 by evening
- Star trust: star growth looks organic
- Category: AI agents · Language: C++ · Created: 2024-10-04 · Last push: 2026-10-02
- GitHub: https://github.com/moonshine-ai/moonshine · Homepage: https://moonshine.ai · Page: https://gitnova.dev/en/r/moonshine-ai/moonshine

## Useful for

- Run microphone speech recognition via the CLI with moonshine-voice mic
- Embed streaming transcription into an iOS, Android, or Windows app
- Build a voice agent with intent recognition and spoken responses without cloud APIs

## Why it’s here

- Star-counter measurements on 2026-11-01 (UTC), 00:07–15:54: 11343 → 11366 stars (+23). This is the change over that interval.
- Estimated end-of-day forecast: about +33 stars, using observed gains and the previous day.
- Over the last two days the pace is 4.7× that of the previous week and a half.
- Elevated gains were measured on the last 2 completed UTC days in a row. Continued growth is not guaranteed.
- GitHub Trending C++ today: #11, +27 stars.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 638
- Issues and pull requests: 216
- Watchers: 64
- Average over the last week: 14 per day
- Usual pace: 6 per day
- Stars in the last hour (measured): 2
- Latest release: v0.1.5 (2026-08-24)

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-10-03 … 2026-11-01: 3, 4, 7, 6, 6, 7, 4, 5, 7, 5, 7, 14, 5, 8, 4, 8, 2, 3, 3, 9, 6, 8, 6, 14, 7, 3, 3, 17, 22, 23

## Magnitude by day, last 2 days

10-31 1.8, 11-01 2.8

## Spotted in now

- GitHub Trending C++ today: #11, +27 stars

## Similar by description

1. **JMS1717/ming-voice** — 0.9 · Cooling · AI agents · Python · +3 stars measured 2026-11-01, 00:08–15:59 UTC, ≈ 4 by evening
   A fully local real-time voice assistant built on the Ming-3.0-audio-tiny speech-to-speech model, running on AMD Radeon GPUs (ROCm/WSL2) with no audio leaving the machine.
   Full card: https://gitnova.dev/en/r/JMS1717/ming-voice.md
2. **andrisgauracs/whistle-esp32** — 0.8 · Steady · Hardware, IoT & robotics · C · +1 star measured 2026-11-01, 00:09–16:00 UTC, ≈ 1 by evening
   An independent C99 implementation of the Whistle speech-to-text model running entirely on an ESP32-S3 microcontroller. Hold a button, speak a colour command in one of seven languages, and an onboard LED responds - with no network, cloud,…
   Full card: https://gitnova.dev/en/r/andrisgauracs/whistle-esp32.md
3. **oxnw/ai-phone** — 3.3 · Cooling · AI agents · Python · +19 stars measured 2026-11-01, 00:03–15:53 UTC, ≈ 29 by evening
   A local SIP phone daemon with speech recognition and a language model on Apple Silicon: it answers and places calls and holds the conversation by voice. Runs on macOS with local MLX models or external LLMs.
   Full card: https://gitnova.dev/en/r/oxnw/ai-phone.md
4. **BoumedineBillal/esp32-pie-asr** — 1.2 · Steady · Hardware, IoT & robotics · C++ · +1 star measured 2026-11-01, 00:08–15:58 UTC, ≈ 2 by evening
   Live English speech recognition running entirely on the ESP32-P4 microcontroller: a 65.7M-parameter Kroko streaming transducer with 2/4-bit weights (26.65 MB) runs faster than real time with no cloud or neural accelerator. Includes six…
   Full card: https://gitnova.dev/en/r/BoumedineBillal/esp32-pie-asr.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-11-01 16:01 UTC, updated every 30 minutes.