LOADING DATASHEET
GROUNDED IN REAL HUMAN CONVERSATIONSYOUR BENCH
Voice Activity Detection
◆ OPEN
by Pyannote
OPEN WEIGHTSLOCAL
RANKED #12 OF 26 AUDIO · OVERALL #302 OF 493 RANKED · VIBE SCORE 5.8 · 109 VOICES ·
COPY BADGE pastes a README snippet with the rank SVG.
voice-activity-detection on Hugging Face (automatic speech recognition). 4,300,717 downloads. Open weights for local or hosted use.
HAPPY9.5
2 mentions
SAD5.5
1 mentions
VIBE TRAJECTORY
26 weeks · 87 voices in this window
HOLDING UP
Recent vibe is steady. No clear fade in the crowd.
HOW TO USE IT
Open weights. You can use a hosted service, or download it and run it yourself, free.
RUN IT ON YOUR OWN MACHINE
Open weights: download Voice Activity Detection once and it's yours. No subscription, no rate limits, works offline.
1.5M parameters (PyanNet architecture), model file is about 6 MB unquantized
- BARE MINIMUM
Intel Core i5-8250U Laptop
8 GB RAM
~$150 USEDFP32 PyTorch on CPU, ~15x realtime speed
- SWEET SPOT
Apple Mac mini (M2, 8-core CPU / 10-core GPU)
8 GB UNIFIED RAM
~$450 USEDFP32 MLX or PyTorch MPS, ~60x realtime speed
- FULL POWER
NVIDIA GeForce RTX 3060 (12GB) Desktop
12 GB VRAM
~$260 USEDFP32 PyTorch on CUDA with batching, ~200x realtime speed
LEARN HOW TO SET IT UP
RUNS INPYANNOTE-AUDIOPYTORCHHUGGING FACE HUB CLI
WILL MY BOX RUN VOICE ACTIVITY DETECTION?
The three cards above are researched picks. Search the machine you actually have — we only say yes if it should reply at a usable speed, not just load the weights and crawl.
HARDWARE PICKS & PRICES ARE RESEARCHED FROM THE LIVE WEB AND REFRESHED AUTOMATICALLY. TREAT THEM AS BALLPARK, NOT GOSPEL.
WHAT CAN YOU ACTUALLY DO WITH IT?
Everyday jobs on each of those rigs: how long Voice Activity Detection takes, and how good the crowd says it is at that kind of work.
| THE JOB | BARE MINIMUMIntel Core i5-8250U Laptop | SWEET SPOTApple Mac mini (M2, 8-core CPU / 10-core GPU) | FULL POWERNVIDIA GeForce RTX 3060 (12GB) Desktop |
|---|---|---|---|
| Narrate a scripta minute of spoken audio | ~4 S★★★★★ | ~1 S★★★★★ | UNDER A SECOND★★★★★ |
| Transcribe a recordinga ten minute recording turned into text | ~40 S★★★★★ | ~10 S★★★★★ | ~3 S★★★★★ |
STARS ARE THE COMMUNITY SCORE FOR THAT KIND OF WORK, DOCKED FOR HOW SQUASHED THE WEIGHTS GET AT EACH TIER. TIMES ARE COMPUTED FROM RESEARCHED SPEEDS. BALLPARK, NOT GOSPEL.
WHAT THE INTERNET SAYS
People are mostly frustrated with Voice Activity Detection right now.
106 POSTS · 4 COMMENTS · DEV.TO · HACKER NEWS · LEMMY · GITHUB · REDDIT · DEV FORUMS · BLOG
PEOPLE LOVE
- Usually gives correct answers (1 mentions)
PEOPLE COMPLAIN ABOUT
- Sometimes makes things up (5 mentions)
TOPIC BY TOPIC
- Getting facts rightA SORE SPOT
1 thumbs up · 5 thumbs down
LOOKING FOR AN ALTERNATIVE?
Other models the crowd has fully reviewed, starting with audio models like this one.
Gemini TTS6.2Google · Audio559 VOICESCOMPARE
ElevenLabs Music> SOURCES (12)
- REDDIT · 2025-12-13 · 109 UPVOTESTTS Audio Suite v4.15 - Step Audio EditX Engine & Universal Inline Edit Tags
TTS Audio Suite v4.15 - Step Audio EditX Engine & Universal Inline Edit Tags
- REDDIT · 2024-06-08 · 70 UPVOTESSillyTavern 1.12.1
SillyTavern 1.12.1
- REDDIT · 2026-06-11 · 42 UPVOTESI built a 100% local, CPU-only voice loop for Ollama — talk to your models hands-free (Silero VAD + Parakeet STT + Supertonic TTS 3)
I built a 100% local, CPU-only voice loop for Ollama — talk to your models hands-free (Silero VAD + Parakeet STT + Supertonic TTS 3)
- BLOG · 2026-09-09 · 32 UPVOTESNVIDIA Brings Real-Time AI to Broadcast, Sports and Global Streaming at IBC
At the IBC conference, running Sept. 11-14 in Amsterdam, the creative, technology and business communities are coming together to turn ideas into action and discuss innovations across the media and entertainment industries. More than 44,000 attendees from…
- REDDIT · 2026-02-28 · 28 UPVOTESFull speech pipeline in native Swift/MLX — ASR, TTS, diarization, speech-to-speech, all on-device
Full speech pipeline in native Swift/MLX — ASR, TTS, diarization, speech-to-speech, all on-device
- REDDIT · 2025-04-02 · 22 UPVOTESReal-Time Speech-to-Speech Chatbot: Whisper, Llama 3.1, Kokoro, and Silero VAD 🚀
Real-Time Speech-to-Speech Chatbot: Whisper, Llama 3.1, Kokoro, and Silero VAD 🚀
- REDDIT · 2025-04-05 · 20 UPVOTESWas looking for open source AI dictation app for typing long prompts, finally built one - OmniDictate
Was looking for open source AI dictation app for typing long prompts, finally built one - OmniDictate
- REDDIT · 2025-07-15 · 16 UPVOTESMy dream project is finally live: An open-source AI voice agent framework.
My dream project is finally live: An open-source AI voice agent framework.
- REDDIT · 2026-07-13 · 8 UPVOTESSelf-hosted voice for any agent/harness of your choice (open-source)
Self-hosted voice for any agent/harness of your choice (open-source)
- HACKER NEWS · 2024-12-23 · 8 UPVOTESShow HN: Open-source turn detection model for voice AI
Show HN: Open-source turn detection model for voice AI
- REDDIT · 2025-07-15 · 8 UPVOTESMy dream project is finally live: An open-source AI voice agent framework.
My dream project is finally live: An open-source AI voice agent framework.
- REDDIT · 2023-09-06 · 7 UPVOTESIntroducing Two Libraries for Voice-Based Applications
Introducing Two Libraries for Voice-Based Applications