FunASR
Verified · 27 days agoTranscribe local audio with FunASR and SenseVoice using private, on-device inference.
Shrink images to a token budget for vision LLM agents and pipelines. Not for chat-pasted images.
The product is real — a Rust image optimizer that snaps images to vision-LLM tile boundaries to cut token costs — but the npm install behavior crosses the line: postinstall silently copies Claude skill files into ~/.claude/skills with no consent gate, mutating an agent-behavior surface in your home directory the moment npm install finishes, and it downloads the platform binary from GitHub releases with no checksum verification. The Claude Code plugin-marketplace path is the consented way to get these skills; the npm path just takes the liberty. Install-time writes to agent config surfaces are a flag regardless of product quality.
Transcribe local audio with FunASR and SenseVoice using private, on-device inference.
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
Turn long videos into viral vertical shorts and publish them to TikTok, Instagram and YouTube.
Trim, convert, resize, compress, and remix audio and video.
Screen recording, meeting notes, and voice dictation - all with AI
Let any LLM watch a video locally — and search everything it has ever watched.