Doza Assist vs Simon Says: for editors cutting interviews on a Mac.
Simon Says is the closest functional comparable to Doza Assist: both transcribe interview footage and both round-trip with Final Cut Pro, Premiere Pro and DaVinci Resolve. The differences are where the transcription runs, how you pay, and what actually comes back into your timeline.
Simon Says is a cloud transcription, captioning and translation service with extensions that run inside Premiere Pro, Final Cut Pro and DaVinci Resolve. Transcripts return as ranged markers, and its Assemble tool lets you drag highlighted lines into a rough cut that exports as a new sequence. Doza Assist transcribes on your Mac, ranks the soundbites, builds a grounded rough cut, learns your style, and places the selects back into the multicam or sync-clip timeline you started from.
01Side by side
| Doza Assist | Simon Says | |
|---|---|---|
| Runs on | Mac with Apple Silicon, 16 GB memory minimum | Web app, Mac app, iPhone app, plus extensions for Premiere Pro, Final Cut Pro and DaVinci Resolve (Mac and Windows). Avid via exported files |
| Where processing happens | On your Mac. Transcription, speaker labels and the language model all run locally. Nothing uploads by default | Cloud. Projects live on the website and audio is sent to third-party speech recognition platforms. A separate on-premise product exists; its Enhanced edition does not run AI on Apple Silicon |
| Pricing model | $199 once with 12 months of updates, then $49/yr if you want more updates. Studio $29/mo or $299/yr | Pay As You Go $15 per hour ($0.25 a minute), or Starter $15, Pro $33 and Pro+ $125 per user per month billed annually with 2, 4 or 30 hours of credit. Extra hours $7.50 down to $4.00 |
| Works inside your FCP, Premiere or Resolve timeline | Reads your Final Cut Pro project (FCPXML 1.11+) and hands selects back into it. Sequences for Premiere Pro (xmeml) and DaVinci Resolve (EDL) | Extensions run inside all three. Transcripts return as ranged markers on clips. Assemble rough cuts export as a new sequence or timeline that then needs relinking |
| Multicam and sync-clip round trip | Yes for Final Cut Pro. Multicam and sync clips come back as the same clip type with the original media references. Premiere and Resolve receive a flat sequence | The FCP extension accepts multicam and synced clips for markers; simple clips recommended. Assemble cannot take in an FCP timeline and return an edited timeline, and Premiere sequence XML does not carry multicam clip structures |
| Speaker labels | Yes, on device, with real names | Auto-speaker identification: you label one row and the AI propagates the name by voice |
| Story selects and rough cuts | Ranked soundbites, Story Brief, Story Builder, and Chat that cannot invent a timecode | Assemble: highlight soundbites and drag them into a rough cut by hand. Ranked soundbites or story suggestions not stated |
| Learns your editing style | My Style (Editorial DNA) learns from cuts you have already finished | Not stated |
| Privacy of footage | Footage never uploads. The optional Claude or OpenAI key is off by default | Machine transcription only. Files retained 6 months after a subscription ends; retention at the third-party speech providers is outside their control |
| Free trial | Every feature on the first 2 minutes of any file. No account | 15 minutes of free credit |
Last checked: September 2026. Simon Says facts from simonsaysai.com and its help center. Where the pricing page and a help article disagree on monthly prices, the pricing page is quoted.
02What each round trip actually does
The phrase round trip covers very different things, so here is each one as precisely as I can state it from the vendors' own documentation and my own code.
- Simon Says and Final Cut Pro. The extension sends clips (including multicam and synced clips) to the cloud and brings the transcript back as ranged markers with speaker labels and favorites, plus captions. Assemble writes an FCPXML that creates a new event with a sequence you relink. Their help center states Assemble cannot bring in a Final Cut Pro project timeline and export an edited timeline back.
- Simon Says and Premiere Pro. Clips from the bin go up, markers come back onto the clips plus a marker sequence. Assemble downloads an XML that builds a new sequence; their help center notes multicam clip structures are not supported in that XML.
- Simon Says and DaVinci Resolve. Markers return color-coded per speaker; Assemble creates a new timeline whose clips need relinking, and multicam works only when the start timecode of the exported file matches the multicam.
- Doza Assist and Final Cut Pro. You drop in the FCPXML of the project you already built. Doza transcribes in timeline time, so every word matches the timecode you cut at. Selects come back as the same kind of clip they came from, a multicam clip stays a multicam clip with its angle, and the file's original resources are spliced in unchanged, so nothing relinks. That frame-exact placement is the part I have a patent pending on.
- Doza Assist and Premiere Pro or Resolve. Selects go out as a Premiere xmeml sequence or a Resolve EDL that references your source media. Those are flat sequences, not multicam round trips, and I say so on every page.
03When Simon Says is the better fit
- Your team is split across Mac, Windows and Avid and needs one browser-based transcript everyone can open.
- The deliverable is captions, subtitles or translation in many languages, or an AI dub, rather than a cut.
- You transcribe a few hours a month and Pay As You Go at $0.25 a minute is cheaper than owning a tool.
- You want real-time collaborative editing of the transcript in a browser.
04When Doza Assist is the better fit
- You want the tool to pick the story, not just transcribe it: ranked soundbites, a Story Brief and a Story Builder that only ever cites real transcript spans.
- You need selects placed frame-exact into an existing multicam or sync-clip timeline in Final Cut Pro, with nothing to relink.
- You want the suggestions to sound like you. My Style learns from cuts you have finished.
- The footage cannot be uploaded, and you need speaker labels and a language model that run on the Mac too.
- You would rather pay $199 once than $15 to $125 per user per month.
05Why not just use the transcription built into your editor?
Final Cut Pro, Premiere Pro and DaVinci Resolve can all turn audio into text now, and for captions that is often enough. Doza Assist is for the step after the transcript, and it adds six things the built-in tools do not:
- Grounded selects. Every soundbite, Story Builder cut and Chat answer points at a real span of the transcript. The app cannot invent a timecode.
- Frame-exact multicam placement. Selects go back into your existing multicam or sync-clip timeline as the same kind of clip they came from, with the original media references intact.
- My Style. Editorial DNA learns how you build a story from cuts you have already finished, so the suggestions sound like you.
- Quote sheets. The transcript becomes a formatted pull-quote document as PDF or Word for producers, clients and comms.
- True on-device privacy. Transcription, speaker labels and the language model all run on your Mac. Nothing uploads by default.
- Pay once and own it. $199 with 12 months of updates, then an optional $49 a year. The app keeps working either way.
Read the full breakdown on Doza Assist vs the transcription built into Final Cut Pro, Premiere and Resolve.
06Frequently asked
Does Simon Says work inside Final Cut Pro?
Yes, through a Mac App Store app that installs as a Final Cut Pro extension. It sends clips to the cloud and returns transcripts as ranged markers with speaker labels, and can send captions into projects. Its Assemble rough cuts export as a new event and sequence rather than into your existing timeline, per Simon Says' own help center.
Which tool keeps footage on the Mac?
Doza Assist runs everything on the Mac: transcription, speaker labels and the language model. Simon Says' mainstream product processes in the cloud with third-party speech engines. Simon Says also sells an on-premise edition; its Standard edition runs on Apple Silicon Macs and its Enhanced edition does not run AI on M-series chips, as of September 2026 per their site.
Can I use both?
Yes. Some editors use Simon Says for caption and translation deliverables and Doza Assist for the story pass and the timeline round trip. Doza Assist's transcript exports as a Word or PDF document, and quote sheets go out the same way.
How does Doza Assist charge for hours?
It does not. $199 once, unlimited hours on your Mac, 12 months of updates included, then $49 a year only if you want to keep updating. Studio is $29 a month or $299 a year.
No account, no card. Transcription, speaker labels, ranked soundbites, Story Builder, My Style and the NLE round trip all run on your Mac.
Promo code SEARCH15 at checkout.
30-day money-back guarantee · Requires a Mac with Apple Silicon (M1 or later) and 16 GB of memory