Hushscript vs YouTubeTLDR
Hushscript
hushscript.comYouTubeTLDR
github.comSide by side
| Figure | Hushscript | YouTubeTLDR |
|---|---|---|
| Price from | $40.83/mo | — |
| Free plan | Yes | — |
| Free trial | — | — |
| Open source | — | — |
| Web | Yes | Yes |
| Windows | — | Yes |
| macOS | — | Yes |
| Linux | — | Yes |
| API | Yes | — |
| Self-hosted | — | Yes |
| Plans published | 8 | 0 |
| Spec score · AI Video Summarizers | 8.2 (#10) | 6.9 (#25) |
| Privacy | For video uploads, audio is extracted in the browser so the video stays on the device; transcription audio is deleted when the transcript is saved, and stored transcripts and Insights are encrypted at rest. | The README says the app runs on the user's own machine and uses their Gemini API key directly, with no proxies or data collection. |
Plans
Hushscript
- 30 free transcription minutesFree
- 45 min$1.99, paid once
- 300 min$5.99, paid once
- 900 min$12.99, paid once
- 1,800 min$19.99, paid once
- Hushscript for Teams$49 a month
- 6,000 min$49.99, paid once
- Hushscript for Teams$490 a year
In their own words
Hushscript
Hushscript has no write-up on Specifiction yet; its figures are above.
YouTubeTLDR
YouTubeTLDR is a free, self-hosted YouTube video summarizer that uses Gemini AI. Users provide a YouTube video URL, choose an available Gemini model and customize the AI’s instructions to shape the summary. The app also provides the full raw transcript, and saves summaries locally in the browser for later reference. Its Rust server runs on the user’s machine; users supply a Google Gemini API key, which the project says is used directly without proxies or data collection. The project describes the application as a roughly 0.3 MB single binary, with no database, Tokio or frameworks, and a vanilla HTML and JavaScript interface. The documented setup is to run the executable and open localhost:8000 in a browser. Server IP, port and worker count can be changed using environment variables. It supports Linux, macOS, Windows, web and self-hosted use, and the repository identifies the license as MIT. The project says it is optimized for personal use, not hundreds of concurrent users. It does not support video uploads or chapter generation.