How to Add Subtitles to a Video Automatically (Free and Offline Options)

A large share of social video is watched with the sound off, and captions make your content accessible to deaf and hard-of-hearing viewers and easier to follow for non-native speakers. Writing subtitles by hand takes 5–10 minutes per minute of video. AI can do the first draft in seconds.

This guide covers the three steps: generate, edit, and attach subtitles to your video.

Step 1: Generate subtitles from speech

YouTube (free, online)

Upload your video (it can be Unlisted). After processing, go to YouTube Studio → Subtitles — YouTube generates automatic captions you can edit and download as SRT. It's free but slow, and your video lives on YouTube's servers.

Clipchamp (Windows 11, online)

Clipchamp, bundled with Windows 11, has auto subtitles in its editor: select the video on the timeline → Captions → Transcribe media. You can export the SRT. Transcription is processed online.

CapCut and phone editors (online)

CapCut and similar apps generate animated captions for short-form video. They're convenient but cloud-based, and exporting a clean SRT isn't always possible on free plans.

Offline AI (private)

OpenAI's open-source Whisper model (and faster reimplementations like faster-whisper) can generate subtitles entirely on your PC. From the command line:

whisper "video.mp4" --model small --output_format srt

See the offline transcription guide for setup, model sizes and hardware. This is the route for confidential footage, unreleased content, or when you simply don't want to upload gigabytes of video.

Step 2: Edit and fix timing

Always review before publishing:

• Names, brands and jargon — AI spells these phonetically.
• Line length — aim for no more than about 42 characters per line and two lines per cue.
• Reading speed — cues should stay on screen long enough to read (roughly 15–20 characters per second).
• Sync — if every line is early or late by the same amount, or drifts over time, the free subtitle converter can shift or stretch all timings at once and convert between SRT and VTT.

Free editors like Subtitle Edit give you a waveform view for precise timing.

Step 3: Attach subtitles to the video

There are three ways, and they suit different platforms:

Sidecar file

Name the subtitle exactly like the video — holiday.mp4 and holiday.srt — in the same folder. VLC, Windows Media Player and most TVs load it automatically. On YouTube, upload the SRT under Subtitles → Add → Upload file.

Embedded track with FFmpeg

ffmpeg -i video.mp4 -i subs.srt -c copy -c:s mov_text output.mp4

This copies the video and audio without re-encoding (it's fast and lossless) and adds a selectable subtitle track. For MKV, use -c:s srt instead.

Burned in with FFmpeg

ffmpeg -i video.mp4 -vf subtitles=subs.srt -c:a copy output.mp4

This re-encodes the video with the text drawn on. To style it, convert the SRT to ASS format or pass force_style, e.g. subtitles=subs.srt:force_style='FontSize=24,Outline=2'.

Accessibility tips

• For closed captions, include meaningful sounds: [laughter], [phone rings].
• Identify off-screen speakers: SARAH: Are you coming?
• Don't cover important on-screen text; move captions to the top if needed.
• Use high contrast — white text with a dark outline or background box.

Doing it all offline, in one app

I built Subtitle Studio AI for Windows to cover all three steps without uploading anything: it generates subtitles with faster-whisper on your PC (a compact model is bundled, with larger models optional), gives you word-level timing and an editable cue table, applies styling presets, and exports SRT, VTT, TXT or ASS — as a sidecar, an embedded track, or burned into a new video, leaving your original untouched. It's a one-time purchase with no account or telemetry.

If you mainly need transcripts of meetings, lectures or interviews rather than subtitles, Whispering Notes AI is the transcription-focused version.