Kotone-DiVE

TTS(Text-To-Speech) bot for Discord, Re-written with golang.
Suitable for self-host usage.

Requirements

• Any computer (including raspberry pi or cloud hosting) that runs Windows / Mac / Linux
  • Linux is our main environment
  • We don't check it, but other every platform golang supports should run it.
• Discord bot account
• Token of your favorite TTS provider
  • Currently supports: azure, gcp, gtts(for testing), voicetext, voicevox, watson, coeiroink, aquestalk-proxy
  • Every single voice type of the provider should work.

How-to

Configure

1. Download config and rename to config.yaml
2. Edit the config file
   • Enter your bot token
   • Enable your TTS provider(s)
   • Disable TTS providers that you don't use
   • Adjust other settings

Setup

• Install docker on your computer
  • Our docker image currently provides: linux/386, linux/amd64, linux/arm64, linux/arm/v6, linux/arm/v7
  • ArchLinux: pacman -Syu --needed docker
• Or if you just want to test or do not want to use docker, simply download binary from releases or GitHub Actions.
  • In that case, make sure that you have installed latest ffmpeg on your computer.
  • Our CI currently builds: linux-386, linux-amd64, linux-arm, linux-arm64, darwin-amd64, darwin-arm64, windows-386, windows-amd64

Run

1. docker run --rm -it -v $(PWD)/config.yaml:/Kotone-DiVE/config.yaml ghcr.io/tpc3/kotone-dive
2. Profit

Tip: database

If you want to save database on docker:

1. Change db path to data/bbolt.db
2. docker run --rm -it -v $(PWD):/Kotone-DiVE/config.yaml -v $(PWD):/Kotone-DiVE/data ghcr.io/tpc3/kotone-dive
3. Profit

Contribution

Contributions are always welcome.
(Please make issue or PR with English or Japanese)