System

Settings

GPU provider, VRAM management, and cloud API configuration.

Voice provider

Local Chatterbox on this PC, or ElevenLabs cloud API.

Offline

Voice provider

Local TTS offline — run start.bat (port 5114) or switch to ElevenLabs in the GPU bar.

Voice AI Service

http://127.0.0.1:5114 · local

Offline

Idle Timeout

Model auto-offloads after 600s idle

600s

No GPU detected

Parallel TTS clips

Chatterbox keeps one model in VRAM — keep this at 1 unless your GPU has enough spare memory to run several clips at once.

Clips at once (1–20)

D:\AI\_VoiceMaster

No models detected

Configure AASeada AI Gateway, Groq, Ollama, and Gemini API pools for video copy generation with 0 local VRAM contention.

Primary AI Provider

AASeada AI Gateway (Remote Multi-Model Cloud)

Offline / Error

Gateway Base URLModel Selection

Gateway API Key (Vault Live Token)

Leave blank to use the built-in system worker key.

Ollama (Local AI Fallback)

Offline (localhost:11434)

Ollama EndpointOllama Model

Pulled from your local Ollama.

Groq Direct API

0 Keys Configured

Groq ModelGroq API Keys Pool (One key per line — rotates automatically on 429 limits)

Google Gemini API

0 Keys Configured

Gemini ModelGemini API Keys Pool (One key per line — rotates automatically on 429 limits)

API keys for cloud voice synthesis

API KeysOne per line. Keys are rotated automatically.

Voice IDModel