skip to content

HOPPOU AI.

Just some cool VRChat stuff we like to make. Voice APIs, browser tools, and Gabriel, an AI who'll actually hang out and talk to you in-world.

projects

07

tools

08

voice api

live

open source

mostly

01 / projects

Stuff we made

02 / api

The voice API

The same voice stack our projects run on, opening up to the public. OpenAI-compatible endpoints, so most SDKs already work with it.

Text to Speech

live

Pocket TTS under the hood. Voice cloning, HTTP and WebSocket streaming, PCM/WAV/MP3/Opus out.

Speech to Text

in testing

Built on parakeet.cpp. Fast transcription on the same base URL, same keys. Public access soon.

# works with the OpenAI SDK

curl https://api.hoppou.ai/tts/v1/audio/speech \
  -H "Authorization: Bearer $HOPPOU_KEY" \
  -d '{
    "model": "pocket-tts",
    "voice": "your-cloned-voice",
    "input": "hello world"
  }' \
  --output speech.mp3

03 / tools

Free tools

Handy stuff for VRChat creators and whoever else. Browser tools run locally, your files never leave your device.