dora/examples/openai-realtime/README.md

• Dora-OpenAI-Realtime (ROOT Repo)
• • Front End
  • • Build Client
    • Run Client
  • Server
  • • Build server
    • Run server
  • On finish
  • GUI
  • • WIP: Moyoyo
  • {Recommended} Install git-lfs
  • Clone Moxin Voice Chat

Dora-OpenAI-Realtime (ROOT Repo)

Front End

Build Client

git clone git@github.com:haixuanTao/moly.git --branch dora-backend-support
cd moly
cargo build --release

Run Client

cd moly
cargo run -r

Server

Build server

uv venv --seed -p 3.11
source .venv/bin/activate
uv pip install dora-rs-cli dora-rs
dora build whisper-template-metal.yml --uv ## very long process

Run server

source .venv/bin/activate
dora up
cargo run --release -p dora-openai-websocket

On finish

dora destroy

GUI

• Go to MolyServer Tab

• Add a custom Provider

• In API Host, use:

  • Name: dora-websocket
  • API Host: ws://127.0.0.1:8123
  • Type: OpenAI Realtime

• Then go to Chat Tab

• New Chat

• ( Make sure the servver is running with: cargo run --release -p dora-openai-websocket)

• On bottom right, click on 🎧 icon.

  If nothing happen is that the server is not found.

• Click on start

• Wait for the first AI greeting

• Start speaking!

• You should get AI response!

WIP: Moyoyo

{Recommended} Install git-lfs

brew install git-lfs # MacOS

Clone Moxin Voice Chat

git lfs install
git clone https://github.com/moxin-org/moxin-voice-chat.git