nowfound

AI · December 19, 2021

Silero VAD

One voice detector to rule them all

Visit site ↗Silero VAD alternatives

What it does

Stellar quality. Highly portable. No strings attached. Supports 8 kHz and 16 kHz. Models < one megabyte in size. Supports 30, 60 and 100 ms chunks. Trained on 100+ languages, generalizes well. One chunk ~ 1ms on a single thread. ONNX up to 2-3x faster.

Does a similar job

all alternatives →
  • IB
    I built a sub-500ms latency voice agent from scratchMar 2026 · ntik.me · ▲570

    I built a voice agent from scratch that averages ~400ms end-to-end latency (phone stop → first syllable). That’s with full STT → LLM → TTS in the loop, clean barge-ins, and no precomputed responses. What moved the needle: Voice is a turn-taking problem, not a transcription problem. VAD alone fails; you need semantic end-of-turn detection. The system reduces to one loop: speaking vs listening. The two transitions - cancel instantly on barge-in, respond instantly on end-of-turn - define the experience. STT → LLM → TTS must stream. Sequential pipelines are dead on arrival for natural…

  • Voxtral2025 · mistral.ai · ▲201

    Frontier open source speech understanding models

  • BY
    Boost Your Voice AI Agents with Open-Source Ten VAD2025 · github.com · ▲8

    Voice Activity Detection (VAD) is a crucial component for Voice AI, enabling more natural and efficient interactions. TEN VAD is an open-source solution designed to supercharge your Voice AI Agents with lightning-fast, human-like conversations! TEN VAD offers some key advantages: ONNX Support: Deploy on virtually any platform or hardware architecture! This means greater flexibility and easier integration into your existing systems. Superior Detection Accuracy: Experience noticeable improvements in voice detection, leading to fewer errors and more reliable performance. Smaller & Faster: Enjoy…

  • Microsoft AI (MAI) Voice-12025 · microsoft.ai · ▲120

    Highly expressive and natural speech generation model

  • MAI-Transcribe-1Apr 2026 · microsoft.ai · ▲85

    Production ASR for noisy multilingual audio

  • Voxtral TTS by Mistral AIMar 2026 · mistral.ai · ▲156

    Multilingual TTS model with realistic and expressive speech

More ai this month

the category →
  • I trained a 125M model to autocomplete piano on-device▲598

    I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes&#x2F;sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.

    AI · 20d ago · simedw.com

  • Astute▲585

    Automate your B2B brand going viral, with new media creators

    AI · 21d ago · company-app.joinastute.com

  • Grok Bot▲547

    AI teammates that you can give real work to

    AI · 28d ago · x.ai

  • Needle2: 14MB agentic LLM for phones, wearables, smart home and robots▲537

    Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens&#x2F;sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens&#x2F;sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…

    AI · 29d ago · cactuscompute.com

  • Monid▲480

    One wallet, every paid tool your agent needs

    AI · 9d ago · monid.ai

  • Clara AI SDR▲460

    Turn website visitors into qualified pipeline

    AI · 22d ago · clarasdr.ai

Source: Product Hunt launch ↗

Launched alongside, December 2021

the whole month →