voice-design - SKILL.md Agent Skill

name: voice-design description: Generate realistic fictional voices across age groups and genders using the official OmniVoice Hugging Face Space. Use when the user asks for natural child, teenager, adult, middle-aged, or elderly voices; male or female voices; voice design; realistic spoken audio; classroom questions; narration; or age/gender voice variants. Use voice cloning only for authorized reference audio.

Voice Design

Use the official OmniVoice Hugging Face Space through the lightweight client. Do not install the local OmniVoice model on this machine: it has an AMD Radeon RX 6750 GRE on Windows, while the supported local GPU paths are not suitable here.

Runtime

Python: C:\Users\Administrator\.local\bin\python.exe
Virtual environment: F:\ai\anything\voice-design-runtime\.venv
Script: C:\Users\Administrator\.codex\skills\voice-design\scripts\generate_voice.py

Generate

& 'F:\ai\anything\voice-design-runtime\.venv\Scripts\python.exe' `
  'C:\Users\Administrator\.codex\skills\voice-design\scripts\generate_voice.py' `
  --text '老师，我有一个问题。' `
  --output 'F:\ai\anything\voice.wav' `
  --gender male `
  --age child `
  --pitch high

Supported controls:

--gender: auto, male, female
--age: auto, child, teenager, young-adult, middle-aged, elderly
--pitch: auto, very-low, low, moderate, high, very-high
--style: auto, whisper
--speed: float, default 1.0

Safety

Generate fictional voices freely.
Clone a real person's voice only when the user confirms authorization.
Do not present generated audio as authentic evidence or as a real person's recording.