Caption generator
Add Sound Effects to Video
This tool adds sound effects to your video automatically, from what is said in it. Upload an MP4 or MOV with speech and CapsAI writes the captions first. AI Sound Effects then reads the caption words and places whooshes, pops, dings, hits and risers from a library of 193 sounds on the moments that need them. You choose a vibe (Auto, Funny, Dramatic or Minimal) and an amount (Subtle, Balanced or Lots), then move, change the volume of or delete any sound on the timeline. The export is one MP4 with the sounds mixed in and the captions burned in. It needs speech, and it picks from the library: it does not generate sounds or add music. 3 free minutes to start, then packs from $4 (₹49 in India, €4 in the euro area), paid once.
Start with 3 free minutes. Watermark-free export comes with every paid pack.
Upload Video
For best results, upload a video with clear speech. You can review and edit the captions after generation.
Drag and drop video or
MP4 or MOV, any aspect ratio.
English subtitle example: This is a preview of the subtitle design Please transcribe to get your video
What the captions look like
Punchline, Funny vibe
And that is how I lost the deposit.
A comic sting such as Sad trombone or Wrong buzzer works best just after the last word, so the line lands first. If it covers the words, nudge it later.
Money line
We made back the whole two thousand in a week.
Cha-ching or Coins on the amount. Both are in the library's money family, and a short one leaves the next sentence clear.
Reveal, Dramatic vibe
Here is the part nobody shows you.
A riser builds under the setup and ends on "nobody". Build-up sounds are placed to finish on the word, not to start there.
List
One, batch your filming. Two, caption everything.
The same light Pop or Click on each number marks the structure. Keep list sounds quieter than hits so they read as punctuation.
What you get
- Sounds land on words, using the caption timings, not a guess from the picture
- Four vibes and three amounts, so a comedy clip and a lesson get different treatment
- 193 sounds in six categories: Transitions, Impacts, Cinematic, Pops & UI, Reactions and Everyday
- Every placement stays editable: move it, change its volume, delete it or add your own pick
- Captions and sound effects leave in one MP4, with nothing to sync in another app
Where it helps
- Talking-head Shorts, Reels and TikToks that need pace without a full edit
- Comedy and storytime clips where a sting after the punchline does the work
- Explainers and tutorials built on lists, numbers and step changes
- Podcast and interview clips that feel flat with captions alone
How it works
How it works
Step 1
Upload a video with speech
MP4 or MOV, up to 10 minutes; files up to 500 MB, or 10 GB with a pack. Choose the spoken language (11 are supported, with no translation). Captions come first because the AI places sounds from the words.
Step 2
Correct the captions
The AI reads the caption text and its word timings. A misheard word can draw the wrong sound, so fix names, numbers and the punchline before adding effects.
Step 3
Choose a vibe and an amount
In the sound panel, set Vibe and Amount, then tap Add AI sound effects. You can cancel while it works; to try other settings later, tap Remove AI sounds and run it again.
Step 4
Review every sound
Tap a sound on the timeline to preview, drag, nudge it in 0.1-second steps, set its volume (0 to 100%) or delete it. Remove AI sounds clears only what the AI placed; sounds you edited stay.
Step 5
Export one MP4
Sounds are mixed into the exported video and the captions are burned in. Post that one file to YouTube, Instagram, TikTok or anywhere else.
Choose your spoken language
What to do next
Related tools
Questions
How do I add funny sound effects to a video?
Upload it, fix the captions, set Vibe to Funny and Amount to Balanced, and tap Add AI sound effects. The AI looks for setups, punchlines and reactions in the words and places stings, pops and reaction sounds there. Play it back, delete anything that steps on a line, and export.
Does it work on a video without talking?
No. AI Sound Effects picks its moments from the captions, and with none the editor says "Add captions first: AI picks the moments from what is said." Music-only clips and silent B-roll give it nothing to read.
Is it free?
A new account gets 3 free minutes of captioning, and the AI Sound Effects panel is in the editor on every account. Exports on the free minutes carry a CapsAI watermark; watermark-free export comes with every paid pack. Packs start at 20 minutes for $4 (₹49 in India, €4 in the euro area), paid once.
Does the AI create new sounds or add music?
No to both. It chooses from the 193 library sounds; it does not generate audio, and CapsAI has no music track or music library. As of 19 October 2026, Captions (captions.ai) advertises a generator that makes sound effects from a text prompt, if that is what you need.
How does it decide where a sound goes?
It reads each caption's words and timings, and the audio is measured for louder words and pauses. Every library sound carries a note on when it fits and when to avoid it. Sounds with a slow start, such as long whooshes, are placed so their loudest point hits the word, and build-ups end on the moment.
How many sound effects can one video have?
Up to 200 per video, with at most 4 playing at once. Amount decides how many the AI places: Subtle for a few accents, Lots for a dense comedy edit.
Can I upload my own sound effects?
No. Only library sounds can be placed. Add a sound of your own in your video editor before you upload.
Will the sounds play on Shorts, Reels and TikTok?
Yes. They are mixed into the MP4, so they play wherever the file is posted. CapsAI keeps your video's shape, so export a vertical cut for vertical feeds. YouTube Help (read 19 October 2026) counts square or vertical videos up to three minutes as Shorts.
Can I download the captions or the sound list as a file?
No. The output is one MP4 with the captions burned in and the sounds mixed in. There is no SRT, VTT or TXT download and no separate audio track.
How long does it take?
A few minutes for a clip under a minute: upload, captioning, the AI pass, then the export render. Listening back usually takes longer than the machine steps.
Which sound goes where: a cue sheet
Cuts and topic changes: a short whoosh or swipe, under a second, peaking on the first word of the new idea. Reveals: a riser or drum roll that builds under the setup and stops on the reveal word, or a shimmer for something new and gentle. Punchlines: a punch for a sharp line, a sad trombone or wrong buzzer for a self-own, placed just after the last word so the joke lands in silence first.
Numbers and money: coins or cha-ching on the figure. List items: the same light pop or click on every item, so the repeat tells viewers where they are. Wins and mistakes: a ding or level-up for a result, a buzzer or glitch for an error. Volume: the AI sets each level from the speech and the sound's intensity; if a sound hides a word, it is too loud. Spacing: one accent per sentence suits a lesson, more suits comedy, and sounds on back-to-back words blur together.
On YouTube
YouTube Studio's Audio Library has a Sound effects tab: you search or filter by category and duration and download each effect as an MP3. YouTube says Audio Library sounds are copyright-safe and are not claimed through Content ID (YouTube Help, read 19 October 2026). Those files still have to be placed by hand in an editor before upload.
After upload, the YouTube Studio editor can add a track from the Audio Library, on a computer and for videos under six hours, but its help page describes adding songs, dragged into place. Neither route puts a sound on a particular word for you. Doing it in CapsAI before upload gives one file with captions and timed sounds; for a Short, keep it square or vertical and up to three minutes, the Shorts rule for uploads since 15 October 2024.
On Instagram
Instagram's own editing app, Edits, lists sound and voice effects and automatic captions in its App Store description (read 19 October 2026). Instagram's Help Center articles on Reels audio would not load for us on 19 October 2026, so we make no claim about the Reels editor inside the Instagram app.
CapsAI does the placing for you in a phone or desktop browser, with no app to install. Export the MP4 and upload it to Instagram as a Reel. If you also add music inside Instagram, play the Reel through once on your phone to check that the music does not bury the effects or the voice.