Reachy Mini · Talk to the robot

Loading Reachy and Microduck

Reachy Mini

Talk to a 3D Reachy Mini, live from your browser — the same conversation experience that ships with the robot!

Tap to start

camera

Conversation

About

This simulator reproduces the conversation app that ships with every Reachy Mini: the same voice pipeline and tools, with a 3D robot standing in for the real one.

View the conversation app on GitHub

Under the hood, the conversation app relies on speech-to-speech: Hugging Face's open framework for real-time voice agents. Rather than one end-to-end model, it chains four open models from the Hub (speech detection, transcription, a vision-language model, and synthesis), so any stage can be swapped or run locally. This demo wires that pipeline to hosted inference.

View the project on GitHub

The pipeline

You speak
VAD detects speech silero-vad
STT transcribes it nvidia/parakeet-tdt-0.6b-v3
VLM analyses and composes the reply · via Cerebras google/gemma-4-31B-it
TTS speaks it back Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice
Reachy replies

Settings

Speech-to-speech server URL URL of your speech-to-speech server, e.g. http://localhost:8080 (the app adds /v1/realtime).

Voice

Microphone Applies on the next conversation (or Restart). Speakers Where assistant audio plays. Chrome/Edge support switching outputs.

Noise gate Off

Off −3 dB

Mutes the mic below the handle so room noise isn't sent. Slide fully left to turn it off; the bar shows your live input while in a call.

Instructions

Connect first, then come back here to apply live changes.

Tools

Let the assistant act during the conversation. Changes apply live.

Web search Current web results from Pollen's DuckDuckGo tool.

Camera Let the assistant see through your webcam.

A live preview shows bottom-left while the camera is on.

Body movement Express emotions and perform dances with Reachy.