Audio generation with gen-ai (CLI & MCP)

Audio is the cheapest modality — typically cents per line — and the fastest to iterate on. This skill covers every audio task you can accomplish via the Picsart gen-ai CLI or the @picsart/gen-ai-mcp MCP server.

When to use

Activate whenever the user asks to:

• Record / generate a voiceover, VO, narration, spoken line, ad read, or podcast intro
• Produce an explainer voice, audiobook snippet, dialogue, or character read
• Localize speech across languages
• Clone a voice or dub an existing line (speech-to-speech)
• Generate music — backing track, jingle, sting, loop
• Generate SFX — UI sounds, whooshes, impacts, ambiences
• Sync audio to a video (video-to-audio / foley)

CLI vs MCP — both work