How to use OpusCut inside your AI

One long video → multiple viral shorts. Great for podcasts, interviews, gaming, livestreams. Source can be YouTube / Google Drive / a direct link; the AI finds the highlights, cuts vertical shorts, and burns captions.

Say

💬“Cut this video into shorts: <video>”

AI asks

Content type (variety / podcast / gaming / tutorial…), caption language, caption style. Spelling it out saves a step.

Output

Multiple vertical 9:16 shorts with burned-in captions; download each when done.

Input

One video per job. A local file works too — attach it in the chat and the AI uploads it first. (To combine multiple clips into one, use ③ Polish.)

Keep the full video, just add captions (translatable, bilingual EN + 中 supported). No cutting.

Say

💬“Add English subtitles to this video: <video>”

Options

Caption language (original / 中 / EN / bilingual / Arabic), caption style, and output ratio (original / 9:16 / 1:1 / 16:9).

Output

The full video with captions (MP4).

Input

One video per job. A local file works too — attach it in the chat and the AI uploads it first. (To combine multiple clips into one, use ③ Polish.)

Add captions + music + AI voiceover + adjust ratio. Key idea: bring in OpusCut, let the AI edit the footage based on what's on screen and in the dialogue; once the rough cut is done, hand it to OpusCut to 'polish' — using beat detection and dialogue analysis to decide how to optimize, add animated captions, and offer AI music to choose from. (Editing is done by a render-capable AI tool such as Claude Desktop / Claude Code; results vary by footage and tool.)

Just answer the AI's questions in this order:

Step 1

Where's the video?

• Already have a URL → give it directly.
• In this chat (uploaded) → ask the AI to upload it to a shareable URL first (e.g. Google Drive shared “anyone with the link”).
• Only raw clips → attach them and tell the AI how to cut; it edits them into one video, then hands it to OpusCut to polish.

Step 2

Narrative or beat-synced?

• Narrative: order clips by story; music added afterwards.
• Beat-synced (cut to the music): pick the music first, then cut to its beats (see Step 4).

Step 3

Music? none / your own track / AI-generated.

AI-generated: the AI makes 3 options for you to preview and pick.

💬“Upbeat vlog music, medium tempo — give me 3 options”

Your own music: the AI can mix it itself, or upload it to a URL for OpusCut to mix (one or the other — avoid double music).

Step 4 (beat-synced only)

Detect the beat: hand the chosen music to OpusCut to detect the beats; it returns the beat timestamps, and the AI cuts/orders shots so cuts land on the beat.

⭐ Step 5 — always preview before finalizing

After the cut, ask the AI to show you a preview before sending to OpusCut for captions or finalizing — don't let it do everything in one shot.

💬“Show me the cut for review before adding captions / finalizing”

For long videos, render a draft quality first to check fast, then the final HD:

💬“Render a draft first for me to check, then the final version”

Step 6

Captions?

• Yes → hand it to OpusCut for captions (music handled at the same time), and set the output ratio (original / 9:16 / 1:1 / 16:9).
• No → if your AI can render, it stitches + delivers directly (OpusCut only does music / voiceover / beats).

Commands worth remembering

How credits work (roughly)

• Clipping / subtitles / polish: scales with video length.
• AI music: charged per option (3 options = 3×); auto-refunded on failure.
• AI voiceover: charged; auto-refunded on failure.
• Using your own music and beat detection: free.

Not connected yet? Back to the connect guide.