Give every small business the voice of a big one.
WaveKat builds open-source, AI-powered solutions that put enterprise-grade capabilities in the hands of everyone. We start with voice — answering phones, handling conversations, being present 24/7 — the kind of thing only big companies could afford to do well. But voice is just where we begin.
| Repo | What it does |
|---|---|
| wavekat-voice | AI phone answering for small businesses — SIP/RTP, voice-to-voice conversations |
| Name | What it does |
|---|---|
| Common Voice Explorer | Browse, filter, and review Mozilla Common Voice dataset clips |
| Repo | What it does | Stars |
|---|---|---|
| wavekat-core | Shared audio primitives (AudioFrame, sample format conversion) |
 |
| wavekat-vad | Voice Activity Detection — WebRTC, Silero, TEN-VAD, FireRedVAD |  |
| wavekat-turn | Turn detection — knows when a speaker is done talking |  |
| wavekat-lab | Interactive dashboard for testing and comparing audio backends |  |
- Framework: Astro (static output)
- Icons: Lucide via
@lucide/astro - Styles: Tailwind CSS
- Deployment: Cloudflare Pages
Requires Node 22 (nvm use 22).
make install # install dependencies
make dev # sync brand assets + start dev servermake build # sync brand assets + build → dist/Logo and wordmark files are sourced from the wavekat-brand submodule — never edited here.
# After cloning, initialise the submodule:
git submodule update --init
# Pull brand updates:
git submodule update --remote vendor/wavekat-brand
# Re-sync assets into public/logos/:
make syncApache 2.0