#6262nd of 88 in VoiceFor voice, speech and music
porcupine
On-device wake word detection engine with SDKs for many platforms
- Stars
- 5k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Dec 2025
- Language
- Python
4,753 on 22 Apr 20265,022 on 11 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Porcupine is a wake word engine that runs a neural network on-device to detect one or several always-listening keywords in a live audio stream. It ships bindings for Python, .NET, Java, Node, C, Flutter, React Native, Android, iOS, web browsers and microcontrollers (Cortex-M, STM32, Arduino), plus Raspberry Pi. Custom keywords are trained in the Picovoice Console, and every SDK requires an AccessKey from it.
Who it is for: Developers adding always-listening wake words to embedded, mobile or web apps
Strengths
- Runs on Cortex-M, STM32, Arduino, Raspberry Pi Zero through 5, mobile and browsers
- Detects multiple keywords with no added runtime footprint
- Official SDKs for about ten languages and frameworks, each with a runnable demo
- Supports nine languages including Mandarin, Japanese, Korean and German
Weaknesses
- Every SDK needs an AccessKey from Picovoice Console, so it is not fully self-contained
- Custom wake words are trained in the hosted Console, not locally
- Built for a few static keywords, not natural-language commands
- Further languages only for commercial customers, case by case
What it needs
- no GPU
- Needs Picovoice AccessKey (Picovoice Console)
- Models: Porcupine keyword files (.ppn), porcupine_params.pv model
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 88| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing, dictation and audiobooks | 98 out of 100 |
| 2 | handyOffline push-to-talk dictation app that pastes transcripts into any text field | 91 out of 100 |
| 3 | whisper.cppC/C++ speech-to-text engine running OpenAI Whisper models locally | 90 out of 100 |
| #4 | buzzOffline desktop app for transcribing and translating audio with Whisper | 84 out of 100 |
| #5 | xiaozhi-esp32ESP32 firmware for a voice chatbot with MCP device and cloud tool control | 83 out of 100 |