FrameVideo Media Preprocessing

Three CLI commands that produce assets for compositions: tts (speech), transcribe (timestamps), and remove-background (transparent video). Each downloads a model on first run and caches it under ~/.cache/framevideo/. Drop the output into the project, then reference it from the composition HTML — see the framevideo skill for the audio/video element conventions.

Text-to-Speech (tts)

Generate speech audio locally with Kokoro-82M. No API key.

Chanjing-Backed Voice Requests

If the user asks for Chanjing voice generation, platform voices, digital-human narration, or any Chanjing-backed speech asset, invoke the chanjing-digital-human skill instead of defaulting to local Kokoro TTS. Only use local npx framevideo tts when the user explicitly wants offline speech or accepts a placeholder while Chanjing auth is unavailable.

npx framevideo tts "Text here" --voice af_nova --output narration.wav
npx framevideo tts script.txt --voice bf_emma --output narration.wav
npx framevideo tts --list                       # all 54 voices

Voice Selection