The following is a deployment configuration intended for Japanese text‑to‑speech using only the VOICEVOX Engine as the backend.
By configuring it appropriately, you can enable other engines as well.
[default_style]
service_id = "VOICEVOX"
style_id = "0"
[tts_services.VOICEVOX.Voicevox]
url = "http://voicevox:50021"
max_chars = 240services:
discord-tts:
image: ghcr.io/yanorei32/discord-tts:master
restart: unless-stopped
environment:
DISCORD_TOKEN: ${DISCORD_TOKEN}
volumes:
- type: bind
source: ./discord-tts.tts.toml
target: /etc/discord-tts.tts.toml
read_only: true
- type: volume
source: data
target: /var/discordtts/
voicevox:
image: voicevox/voicevox_engine:cpu-latest
restart: unless-stopped
volumes:
data:DISCORD_TOKEN=ush9Zohzie6ahmohsoo6meCh.IThah7.jeephaijiachu8kuWoh0aephe5e
https://github.com/yanorei32/discord-tts/wiki/Command-Reference
- VOICEVOX / AivisSpeech
- Native VOICEVOX API support
- OmniVoice
- Native OmniVoice API support
- VOICEROID
- works with https://github.com/yanorei32/aitalked-server
- mirae-tts
- works with https://github.com/yanorei32/mirae-tts
- kttsproject
- works with https://github.com/yanorei32/libktts-server
- WinRT
- macOS say
- works with https://github.com/yanorei32/say-server
- Android TTS (TextToSpeech API)
- ⚠ Google Translate
- based on https://github.com/zlargon/google-tts by @sim1222
- ⚠ NAVER CLOVA
- based on https://github.com/scottgigante/NaverTTS by @REO2248
- ⚠ Bing Speech
- based on https://github.com/rany2/edge-tts by @REO2248
- ⚠ CoeFont Try
- created by @REO2248
- ⚠ CapCut
- works with https://github.com/kuwacom/CapCut-TTS
- ⚠ Volcengine Translate
⚠ Non-official API. No warranty and not supported. Use at your own risk!