All features
Living recordings
Recordings that stay useful.
A recording you can never find again is dead weight. vidspawn transcribes each one on your machine with whisper by default, then keeps the library alive: searchable by what was said, separated by speaker, summarized and chaptered by a local model, your own CLI or your own key.
On-device whisper, by default
Transcription runs over the finished file on your own machine. No cloud, no upload, no API key required. Your own cloud key is the optional fast lane, never the default.
Speakers, separated
Optional on-device diarization separates who said what. You name each speaker once and the names persist.
Search that lands on the moment
Keyword and semantic search fused together. A result deep-links to the exact moment in the clip, not just the file.
Titles, summaries, chapters
An optional downloadable local model drafts titles, summaries, timestamped chapters and tasks for each recording.
Three brains, zero of them ours
Enrichment runs on the embedded local model by default. Prefer your own stack? Point it at the agent CLI you already pay for (claude, codex or gemini), or at your own API key: Anthropic, OpenAI or any compatible endpoint, LM Studio and ollama included.
From public reviews of other recorders
"Id like more flexibility to organize all recorded videos. Sometimes I get a bit lost when searching for past material."
Vidyard user · Capterra ↗"organization can get messy if your workspace grows without naming rules or folders"
Loom user · Capterra ↗Public reviews by users of other products, quoted verbatim and linked. They describe those products, not vidspawn.
"That call where we discussed the migration" is a query, not an archaeology dig.