Most screen recorders either charge monthly for AI features or send your audio to the cloud for processing. DalVideo, a new Windows-only tool, runs its AI features entirely on your machine.
The headline capability is automatic captioning powered by OpenAI's Whisper model, supporting 99+ languages with no internet connection required. Whisper is the same speech recognition model that powers transcription in tools like Descript - but here it runs locally, which matters if you are recording anything confidential.
What's Included
- AI captions: Offline, Whisper-based, 99+ language support
- AI noise removal: Real-time processing using RNNoise
- Smart Trim: Automatically detects and cuts silent sections
- Built-in editor: Trim, split, and delete without switching to another app
- Recording: Up to 4K at 120fps with hardware encoding (NVENC, AMF, QuickSync)
- Codecs: H.264, H.265, AV1 in MP4 or MKV containers
- Extras: Drawing overlay and keystroke display for tutorials
The built-in editor is the practical differentiator. Recording a quick walkthrough usually means exporting to a separate editor just to trim the beginning and end. Having trim, split, and silence removal in the same tool cuts that step out entirely.
DalVideo is free, though the project is early-stage. Windows-only limits the audience, and there is no mention of Mac or Linux support on the roadmap. The offline-first approach is the right call for screen recording - nobody wants their internal tool demos or client calls uploaded to a third-party server for captioning. If you record tutorials, demos, or training videos on Windows and have been cobbling together OBS plus a separate editor plus a captioning service, this consolidates three tools into one.