Presets built for multi-speaker dialogue
Conversational clips move fast, so viewers need three to five words on screen at once with an unmistakable marker on the active word. Karaoke shows up to five words across two lines with a moving purple box behind the spoken word; Bold sets three uppercase words with a yellow active highlight; Boxed places dark text on a solid white card that reads cleanly over split-screen seams.
Some previews use a frame from Sprite Fright by Blender Studio, CC BY 4.0, cropped.
Handling two speakers and guest names cleanly
When two hosts or a guest trade quick lines, you never want the end of Speaker A's sentence sitting inside the same caption block as Speaker B's reply. Select the last word before a speaker switch and press B: Cuecaps forces a clean caption break so the new voice starts a fresh line.
Guest names, company names and niche terminology are where generic auto-caption tools fail. Before running Cloud transcription (ElevenLabs Scribe v2), paste your guest's name and key terms into the vocabulary box so they come back spelled right the first time—or fix any recurring word across the whole clip with Find and Replace.
Framing 16:9 studio cameras into 9:16 vertical clips
If your podcast is filmed in 16:9, set Shape to 9:16 under Layout: choose Fill frame to crop in on a single speaker (dragging or zooming the canvas to center their face), or Fit inside to letterbox the wide shot. If you edit stacked two-camera split-screens in Premiere Pro or DaVinci Resolve, drag your captions to 50% height so they sit right on the horizontal seam between the top and bottom camera angles.
- Export your vertical cut (or audio track) from your editor and drop it into Cuecaps.
- Click Find words and run Cloud transcription (best for two voices, room reverb and crosstalk) with your guest's name in the vocabulary box.
- Press B at speaker handoffs and press E on the punchline or hook word you want emphasised.
- Export a 9:16 MP4 for Shorts, Reels and TikTok, a transparent MOV overlay for your NLE, or an SRT for the full episode on YouTube.
Questions
How do I keep two speakers' lines from merging into one caption?
Cuecaps transcribes every spoken word sequentially on the timeline. Whenever the speaker changes, click the last word before the switch and press B (Break)—the next speaker's first word immediately starts a new caption group.
Can I transcribe a full two-hour podcast episode for YouTube SRTs?
Yes. Cloud transcription accepts single files up to four hours long, and on-device transcription has no duration limit. You can export a full-episode SRT and VTT file from the same studio.
Does my multi-gigabyte podcast video have to upload?
Never. Your video stays on your disk; cloud transcription sends only compressed speech audio and deletes it once word timings return. The Free plan includes 10 cloud minutes a month plus unlimited on-device transcription.
Try it on your video
You can also drop a file anywhere on this page. Or tryTry the demo clip