diff --git a/config/docker-compose.yml b/config/docker-compose.yml index a9980c1..2ef0639 100644 --- a/config/docker-compose.yml +++ b/config/docker-compose.yml @@ -1,7 +1,25 @@ +# ───────────────────────────────────────────────────────────────────────────── +# Qwen3-TTS GPU Stack +# +# Ports: +# 8020 -> Voice clone, OpenAI-compatible /v1/audio/speech +# 8021 -> VoiceDesign model for prompt-designed voices +# 8022 -> CustomVoice model +# 8023 -> Low-latency streaming voice clone endpoint +# +# Voice files: +# The voice clone and streaming services scan active voices from: +# /home/sparky/Projekte/TTS_Voices/active_voices +# +# Portainer note: +# Set HF_TOKEN in the stack environment if the models ever need Hugging Face access. +# ───────────────────────────────────────────────────────────────────────────── + services: - faster-qwen3-tts: + # Standard voice cloning backend used by TTS Voice Creator and other OpenAI-compatible clients. + faster-qwen3-tts-voiceclone: image: faster-qwen3-tts-dgx-spark:v4 - container_name: faster-qwen3-tts + container_name: faster-qwen3-tts-voiceclone restart: unless-stopped runtime: nvidia environment: @@ -32,6 +50,7 @@ services: networks: - dgx_net + # VoiceDesign backend: creates speech from a descriptive voice prompt instead of a reference clip. faster-qwen3-tts-voicedesign: image: faster-qwen3-tts-dgx-spark:v4 container_name: faster-qwen3-tts-voicedesign @@ -63,6 +82,7 @@ services: networks: - dgx_net + # CustomVoice backend: keeps the separate CustomVoice model available on port 8022. faster-qwen3-tts-customvoice: image: faster-qwen3-tts-dgx-spark:v4 container_name: faster-qwen3-tts-customvoice @@ -94,9 +114,10 @@ services: networks: - dgx_net - qwen3-tts-streaming: + # Streaming voice clone backend: same active voices as 8020, but streams WAV chunks while generating. + faster-qwen3-tts-streaming: image: qwen3-tts-streaming-dgx-spark:latest - container_name: qwen3-tts-streaming + container_name: faster-qwen3-tts-streaming restart: unless-stopped runtime: nvidia environment: