Skip to content
SPECIFICTION
Get listed

Auto Subtitle

github.com

WebWindowsmacOSLinuxSelf-hosted
Visit github.com
Spec score 7.3 #93 of 99 in AI Video Caption Generators

Spec sheet

Purpose
Auto Subtitle is a local-first browser workspace that generates, reviews, edits, and exports subtitles from video.Read 12 Oct 2026 · github.com
Product
Auto Subtitle is a local-first browser workspace for generating, reviewing, and exporting subtitles from video.Read 12 Oct 2026 · github.com
Free plan
NoRead 19 Sep 2026 · github.com
Support
The repository says issues and focused pull requests are welcome, and recommends opening an issue before larger changes.Read 12 Oct 2026 · github.com
License
The project source code is MIT licensed.Read 12 Oct 2026 · github.com
Burn-in captions
NoRead 19 Sep 2026 · github.com
Caption styling
NoRead 19 Sep 2026 · github.com
Debug logs
Exported debug logs can include recognized text and timing details but do not include audio or video bytes; the guide advises reviewing logs before sharing.Read 12 Oct 2026 · github.com
Editing
The editor includes a video player with caption overlay, a magnetic timeline, and tools to edit, split, merge, retime, and validate subtitle cues.Read 12 Oct 2026 · github.com
File formats
It imports SRT, WebVTT, and Auto Subtitle project JSON, and exports SRT, WebVTT, TXT transcripts, and project JSON.Read 12 Oct 2026 · github.com
Import and export
It imports SRT, WebVTT, and Auto Subtitle project JSON, and exports SRT, WebVTT, TXT transcripts, and project JSON.Read 12 Oct 2026 · github.com
Languages and models
The guide lists Auto with English fallback, English, Spanish, French, German, Japanese, Korean, and Chinese, plus Tiny, Base, Large v3 Turbo, and Distil Large v3 models.Read 12 Oct 2026 · github.com
Limitations
Transcription can take significant memory and time, the complete decoded audio buffer is held in browser memory, and videos with multiple audio streams use the first stream.Read 12 Oct 2026 · github.com
Local data
Project autosave uses browser IndexedDB, while theme preference and diagnostic history use localStorage.Read 12 Oct 2026 · github.com
Local data and privacy
The project says video and audio are processed in the browser, project autosave uses IndexedDB, and the app has no account system, application backend, analytics, or external AI API integration.Read 12 Oct 2026 · github.com
Local transcription
It runs Whisper ONNX models through Transformers.js in the browser and extracts audio with FFmpeg.wasm.Read 12 Oct 2026 · github.com
Model downloads
Network access is needed to install dependencies and normally to download each selected model the first time; cached models can run locally afterward.Read 12 Oct 2026 · github.com
Network use
Network access is needed to install dependencies and normally to download each selected Whisper model the first time; later transcription can run locally while the model remains cached.Read 12 Oct 2026 · github.com
Operating requirements
The README lists 64-bit Windows, macOS, or Linux with a current desktop browser, Node.js 20.19+ or 22.12+, and 8 GB RAM as a practical minimum.Read 12 Oct 2026 · github.com
Privacy and integrations
The project states that it has no application backend, accounts, analytics, tracking, paid transcription API, or external AI API integration; selected media is processed in the browser.Read 12 Oct 2026 · github.com
Range regeneration
Users can regenerate selected subtitle ranges up to 29 seconds and compare multiple local alternatives.Read 12 Oct 2026 · github.com
Requirements
The README lists 64-bit Windows, macOS, or Linux with a current desktop browser, Node.js 20.19+ or 22.12+, and recommends at least 8 GB RAM, with 16 GB or more recommended for typical use.Read 12 Oct 2026 · github.com
Subtitle export
other formatsRead 19 Sep 2026 · github.com
Supported video
Supported video inputs are MP4, WebM, MOV, and MKV, with codec support depending on the browser and FFmpeg.wasm.Read 12 Oct 2026 · github.com
Transcription
It runs Whisper models through Transformers.js in the browser and extracts audio with FFmpeg.wasm.Read 12 Oct 2026 · github.com
Translation limit
Large v3 Turbo and Distil Large v3 are transcription-only in the app; Distil Large v3 is English-only and provides segment timing.Read 12 Oct 2026 · github.com
Usage limits
The guide warns for videos larger than 500 MB or longer than 30 minutes, and says browser transcription can be memory intensive.Read 12 Oct 2026 · github.com
Video inputs
Supported video inputs are MP4, WebM, MOV, and MKV.Read 12 Oct 2026 · github.com

Where Specifiction places it

Alternatives to Auto SubtitleAll alternatives →