AI clip maker for musicians: turn any video into viral shorts
Paste a YouTube link or upload a video. TONTRIA transcribes it, finds the best 30–90 second moments, reframes them to vertical following whoever is singing or talking, and adds animated karaoke captions — ready for TikTok, Reels and Shorts.
1 credit = 1 minute of video processed. The free plan includes 30 credits to start, no card needed.
How to make viral clips with AI in four steps
The process is the same whether you start from a two-hour livestream, a podcast appearance or a three-minute acoustic take.
- Paste the link or upload the video
Any YouTube video you own or have permission to use: a live session, an interview, a music video, a podcast episode. You can also upload the file.
- Choose an edit mode and a caption style
Pick viral – spoken for talking, viral – sung for performances, then one of the caption templates. Your brand preset (fonts, colors, logo) is applied automatically if you saved one.
- Review the clips the AI suggests
Each clip comes with a virality score and its transcript. Open the strongest ones first, adjust the start or end, and fix any word in the captions.
- Export for every platform
Download the vertical 9:16 version for TikTok, Reels and Shorts, or open the clip in the editor to create 1:1, 4:5 and 16:9 versions too.
Good source material
- Live sessions and acoustic versions — one take can become five or ten clips.
- Interviews, radio and podcasts — stories behind the songs travel well on social.
- Livestreams and Q&As — hours of content nobody will watch in full.
- Music videos and behind the scenes — vertical cut-downs for release week.
Free AI clip maker: what each plan includes
Every plan includes AI Clips. 1 credit = 1 minute of video processed, and every clip found in that video is included, so a 30-minute live session uses about 30 credits.
| Plan | Credits | Video you can clip | Max export | Watermark | Fonts |
|---|---|---|---|---|---|
| Free | 30 once, no card | About 30 minutes | 720p | TONTRIA watermark | 20 |
| Creator — USD 10/month | 150 per month | About 2½ hours a month | 1080p | No | 220+ |
| Pro — USD 20/month | 300 per month | About 5 hours a month | 4K / 60 fps | No | 220+ |
| Label — USD 99/month | 1,500 per month | About 25 hours a month | 4K / 60 fps | No | 220+ |
From a live show to a sung clip
This clip was picked automatically from a 12-minute live show. The “viral – sung” edit keeps the singer in frame and the karaoke captions follow every word.
Everything a good short needs, done for you
Clipping by hand means scrubbing through hours of footage, cropping frame by frame and typing captions. These are the parts TONTRIA automates.
Word-level transcription
The whole video is transcribed word by word, with the language detected automatically — Spanish, English, Portuguese and many more. That transcript drives both the moment selection and the captions.
Best moments with a virality score
The AI proposes clips between 30 and 90 seconds and rates each one, so you know where to start instead of watching everything.
9:16 reframe that follows the face
The vertical crop follows the person who is talking or singing, instead of a fixed center crop that cuts people out of frame.
Edit modes for music
Viral – spoken follows the conversation. Viral – sung stays on the singer and cuts to the reactions around them.
Animated captions
Karaoke, word by word, viral and phrase templates. Any font, color, stroke and shadow, so captions match the artist instead of looking generic.
Logo and brand presets
Add a logo overlay and save fonts, colors and caption style as a preset. Every new clip starts on-brand.
Why clipping music is different from clipping a podcast
Most AI clipping tools were designed for people talking to a camera. That works for a podcast, but a song is not a conversation. A chorus does not have a “key sentence”, and a good performance clip depends on the voice, the energy of the room and the faces of the people listening.
TONTRIA was built inside the music business, so it treats both cases differently. In spoken content the AI looks for complete ideas, punchlines and stories. In sung content the viral – sung edit keeps the singer in frame while the voice carries the clip, and cuts to reactions so the short has movement. The karaoke captions highlight each word as it is sung, which is what makes people sing along in the comments.
Use cases for artists
- Turn a live session into a week of daily posts before and after a release.
- Cut the best stories from an interview or a podcast appearance with captions that keep people watching without sound.
- Make vertical versions of the music video for TikTok, Reels and Shorts.
- Recycle old livestreams and shows into new content between releases.
Use cases for labels and managers
- Process press days and showcases for several artists without a dedicated editor.
- Keep each artist’s captions, fonts and logo consistent with brand presets.
- Reactivate catalog: clips from archived performances can bring new listeners to older songs.
Tip: post several clips from the same session over a few days rather than all at once. Different moments reach different people, and you learn which part of the song connects.
Want to fine-tune a clip, add text or export square and horizontal versions? Open it in the video editor. Need a lyric video for the same song? Use Video Lyrics.
AI clips: frequently asked questions
How does TONTRIA choose the best moments of a video?
First it transcribes the whole video word by word. Then the AI reads the transcript and looks for moments that work on their own as a short: a strong hook in the first seconds, a complete idea or chorus, and a clean ending. Each suggestion is 30 to 90 seconds long and comes with a virality score so you can start with the strongest ones. You can always adjust where a clip starts and ends.
Can the vertical crop follow the singer instead of staying in the center?
Yes. When it reframes a horizontal video to 9:16, TONTRIA follows the face of the person who is talking or singing. In the viral – sung mode it also cuts between the singer and the reactions of the people around them, the way music clips usually look on TikTok and Reels.
What is the difference between the “viral – spoken” and “viral – sung” modes?
Viral – spoken is for interviews, podcasts, behind the scenes and anything where people talk: the edit follows whoever is speaking. Viral – sung is for live sessions, acoustic versions and performances: the edit stays with the voice and adds cuts to the reactions, so the clip feels alive.
Which caption styles can I use?
There are four animated templates — karaoke, word by word, viral and phrase — and you can change the font (20 on Free, 220+ on paid plans), color, stroke and shadow. Captions are generated from the transcript, so you only need to fix the occasional word.
Is there a free AI clip maker for musicians?
Yes. TONTRIA Clips has a free plan with 30 credits to start (1 credit = 1 minute of video processed), no card needed, so you can clip a 30-minute live session or several interviews before paying. Unlike general AI clipping tools built for podcasts, it has a viral – sung edit that stays on the singer and karaoke captions that follow every sung word. Paid plans add more credits each month and more fonts.
Can I make AI clips without a watermark?
Clips exported on the free plan carry a TONTRIA watermark and go up to 720p. Any paid plan removes the watermark: Creator (USD 10 a month) exports in 1080p, and Pro and Label up to 4K at 60 fps, with 9:16, 1:1, 4:5 and 16:9 versions at once. See the pricing page.
Does it work with livestreams?
Yes, once the stream has ended. The step-by-step for YouTube, Twitch and TikTok streams is on the live show clips page.
Can I upload a video instead of pasting a YouTube link?
Yes. You can paste a YouTube link or upload the file directly. Use content you own or have permission to use.
How many credits does a clip project use?
One credit equals one minute of video processed. A 20-minute live session uses about 20 credits, and every clip it produces is included. Editing and re-exporting afterwards does not use credits.
Keep exploring
Turn your next video into clips
Paste a YouTube link and see the clips TONTRIA suggests.
The free plan includes 30 credits to start, no card needed.