Sub for Creator 繁中 View on GitHub

OPEN-SOURCE AI SUBTITLES

Turn every spoken word into captions people can see.

Sub for Creator is a free, open-source subtitle workspace for creators. Upload a video, let AI transcribe and segment the speech, then fine-tune every moment on the timeline.

See how it works Explore the source

NO ACCOUNT · NO MODEL TRAINING · AGPL-3.0

episode-07.mp4 READY
00:01:24 / 00:12:48
Let your best work be seen.every word matters
00:0000:0400:0800:12
Let your bestwork be seenevery word matters
✦AI transcriptionWord timestamps ready
✓Export readySRT · ASS · MP4
01AI transcription
WhisperX / faster-whisper
02Timeline editing
Word alignment, precise control
03Flexible export
Captions, renders, alpha video
04Open by default
AGPL-3.0 · self-hostable

WHY SUB FOR CREATOR

From voice to final cut,
subtitle work stays simple.

No more jumping between tools. Transcribe, review, style, and export in one focused workflow.

01 / TRANSCRIBE

AI transcription,
word by word.

WhisperX creates precise word-level timestamps, so captions do not just contain the right words—they arrive at the right moment.

00:01.20Every sentence00:02.86has its own timing00:04.12
02 / EDIT

Edit every moment
on the timeline.

The player, subtitle list, and wavesurfer timeline stay in sync while you edit text and timing with an immediate preview.

03 / EXPORT

One transcript,
every destination.

Send a caption file to your editor, burn subtitles into a social clip, or keep working in your NLE.

.SRT.VTT.ASS.MP4.WEBM
04 / PRIVATE BY DEFAULT

No account required.
Your content stays yours.

Anonymous sessions, no names or emails, automatic deletion after 48 hours, and no training on your uploads.

48hautomatic deletion0profile dataAGPLopen license

A SIMPLE WORKFLOW

Three steps to a captioned video.

No subtitle expertise required. Start with the common path and refine as much as you like.

01

Upload a file

Drop a video or audio file into the upload area, then choose a language.

MP4 · MOV · MP3 · WAV
02

Let AI process it

Transcription, punctuation, and smart segmentation run in a visible queue with live progress.

QUEUE → TRANSCRIBE → ALIGN
03

Review and export

Adjust words, timing, and style in the editor, then choose the output that fits your destination.

EDIT → PREVIEW → EXPORT

MADE FOR CREATORS

One source,
every way to publish.

Whether you are posting to social media, handing work to an editor, or continuing in your NLE, export the format that matches the next step.

Read the full documentation ↗
Caption files
SRTVTTTXT
Players and platforms
Advanced
ASSFCPXML
Styles, karaoke, editing workflows
Video output
MP4WEBM
Burned-in or transparent video
✦ASS export supports per-word karaoke color changes for lyrics and sing-along videos.

QUICK START

Want to self-host?
Start in minutes.

OPEN
SOURCE
01

Deploy with Docker

Use this path to start the complete service stack. You need Docker, Redis, and a host that can run speech recognition.

TERMINAL
cp .env.example .env docker compose up -d --build # Open http://localhost:8080
02

Develop locally

The frontend uses React + Vite and the backend uses FastAPI + Celery. Run the API, worker, and web app independently.

TERMINAL
git clone https://github.com/nashwu0117/Sub-For-Creator.git cd Sub-For-Creator npm --prefix frontend install npm --prefix frontend run dev
Have a GPU? Use WhisperX. CPU-only hosts can switch to faster-whisper.Read README

PRIVATE BY DEFAULT

Your footage is your work.

Anonymous sessions, no model training, and automatic deletion after 48 hours. Read the privacy policy and terms of service.

BUILT IN THE OPEN

Make subtitles easier to do well.

This is a growing open-source project. Report a bug, share an idea, or send a pull request.

Open an issue ↗View AGPL-3.0