Build a Repeated-Takes Comparison Sequence

This workflow turns one or more selected recording clips into a review-friendly comparison sequence.

Instead of scrubbing through a long interview or line-reading session and manually sorting duplicate deliveries, Automation Agent analyzes the existing source transcripts, identifies repeated intended lines, and builds a new sequence where each line appears as one stacked take group.

New to Automation Agent?

This recipe uses Automation Agent for Adobe Premiere. Start with Set Up Your AI if the panel is not installed yet. The ready-made workflow itself uses deterministic local transcript analysis and does not require a configured Workflow AI provider.

Offline deterministic workflow

The finished Build Repeated-Takes Comparison Sequence workflow runs locally from Library > Examples > Editing.

• Works offline: transcript analysis and sequence planning stay on your computer.
• No agent setup: it does not require Workflow AI, MCP, an API key, or a configured agent provider.
• No token costs: repeated-take matching is deterministic rather than model-based.
• Seconds instead of minutes: typical short sessions finish in a few seconds and may feel nearly instant, while the live-agent approach can take several minutes. Exact times depend on transcript length and hardware.

The Python Runtime Pack must be installed, and every selected source must already have a usable Premiere transcript. Once that runtime is available, the workflow is ready to run without any additional AI configuration.

Fastest Path: Run The Library Workflow

1. Open Library > Examples > Editing and run Build Repeated-Takes Comparison Sequence.
2. While the setup dialog is open, directly select or adjust one or more recording clips from the same session in the Project panel. Do not select a bin or sequence.
3. Confirm that every selected clip is online and already has a usable transcript. At least one source must contain video; additional audio-only recordings are accepted.
4. Optionally paste the intended script or transcript.
5. Adjust the natural before/after handles or the gap between take groups if needed, then click OK.

Providing the intended text is the preferred mode: matching passages are placed in script order, and repeated takes of the same passage are stacked on separate tracks. One source clip can contribute several separate takes; the workflow does not treat the complete recording as one indivisible match.

If you leave the field empty, the helper conservatively reconstructs repeated wording from the selected source transcripts. This works well for repeated or lightly changed deliveries, but it does not claim to understand semantically equivalent free-form paraphrases.

Video Walkthrough

Short walkthrough of the rough-select comparison-sequence workflow in Premiere.

When This Is Useful

Use this workflow when:

• one or more recording clips contain repeated deliveries of the same lines
• you want to compare alternate takes side by side in timeline form
• you want dead air and obvious waiting sections removed before review
• you want later takes on V1/A1 with earlier alternatives parked on muted higher tracks

Typical examples:

• presenter pickups
• piece-to-camera line variations
• interview answer retries
• social promo lines recorded several times in one long take

What The Library Workflow Creates

The result is a new non-destructive review sequence:

• each intended passage occupies one timeline group
• alternate deliveries are aligned around shared spoken words and stacked vertically
• takes are ordered chronologically, with the latest take on the lowest lane
• higher video and audio lanes are muted initially
• small handles preserve natural context around each delivery
• file boundaries remain hard take boundaries, even when the recording was stopped and resumed
• sequence markers identify missing reference passages, accepted fuzzy/partial stacks, or excluded plausible ambiguities
• with a supplied reference, owned source markers label each placed take as EXACT, WORDING, or PARTIAL

The workflow does not call a take “best” or rank performance quality. The latest take is simply placed lowest because presenters often repeat a passage until satisfied.

If you already have the desired final script or paper edit and want Premiere to assemble only that text in its specified order, use Assemble Sequence From Script (Offline) instead. This Repeated-Takes workflow is exhaustive by design: it organizes the repeated deliveries for review rather than treating a target text as the edit blueprint.

Permissions For The Library Workflows

Both Library workflows on this page create a new sequence and place ranges from the selected source clips. If you restrict Project write access, allow both the selected source items and the Premiere Project-panel location where the new sequence will be created. The workflows use Premiere's default location and have no output-bin picker, so allowing a separate output bin alone is not sufficient.

Source access is needed because placement temporarily sets and restores source In/Out points. Build Repeated-Takes Comparison Sequence can also add its own AA-RT review markers to source items when you provide a reference script. The workflows do not change the underlying media files.

Both workflows need permission to run the installed Python Runtime Pack and to read, write, and clean up their working folders under ::HOME::/automation-agent-repeated-takes or ::HOME::/automation-agent-semantic-rough-select. The deterministic workflow does not require internet access. Create Semantic Rough-Select Comparison additionally sends transcript text and your optional editing instructions to the configured Workflow AI provider, subject to that provider's privacy, network, and token settings.

Source Order

When all selected files contain coherent capture-date metadata, the workflow uses that chronology. Otherwise it keeps Premiere's selection-API order for the complete set and reports the fallback. A file boundary always starts a new take.

Semantic Free-Speech Library Workflow

Use Create Semantic Rough-Select Comparison from Library > Examples > Editing when the recordings are freely spoken and alternative passages may express the same idea with substantially different wording.

This Workflow AI variant gives the configured provider the complete transcript context as stable text-unit IDs. The provider proposes an ordered story structure and identifies local alternative constructions; deterministic code then validates those IDs, resolves timing from Premiere's transcript data, and builds the sequence. The provider never authors timestamps, track indices, or Premiere references.

To run it:

1. Open the workflow and optionally enter an outline or special editing instructions. For example, explain whether deliberate dramatic pauses should be preserved or tightened, or whether obvious false starts should be removed aggressively for a tight social edit.
2. Adjust the natural before/after handles or the gap between content blocks if needed.
3. While its setup dialog remains open, directly select one or more online, already-transcribed footage or audio clips in the Project panel.
4. Click OK and allow Workflow AI to analyze the complete recording session.
5. Review every resulting block, alternative, gap, marker, and omission report.

The workflow gives Workflow AI a sparse indication when significant silence occurred between transcript units. Silence is context, not an automatic cut: the agent must still distinguish a restarted take from reflection, hesitation, or an intentional dramatic pause using the wording, complete session, and your optional instructions.

This mode intentionally does not select one winning edit:

• required singleton passages form the main path on V1/A1
• mutually exclusive local constructions are stacked on higher tracks
• the next story block begins after the longest local alternative
• track meaning resets for every block, so V2 in one block is unrelated to V2 in the next
• unresolved material can appear in a separate review area
• omitted and reused transcript units are reported rather than prohibited
• uncertainty, conflicts, qualifications, and refined boundaries can produce review markers

Editorial coherence and clean connections take priority when the agent orders local alternatives. When two constructions are otherwise similarly useful, it prefers the one using the later recorded take on the lower track. This is a practical review heuristic, not a claim that the later performance is better. You can override it in the optional editing instructions for projects with a different take-selection convention.

Provider, privacy, and review requirements

This workflow requires a configured Workflow AI provider with JSON-Schema structured output. Transcript text and optional editorial context are processed according to that provider's configuration and may incur network use, token costs, and several minutes of processing time. Input-size profiles are advisory rather than hard limits: a capable provider may accept larger sessions, while an overloaded provider may fail without changing Premiere.

The result is an editorial proposal, not a verified final story. Original footage remains available, and a human editor must review material the proposal omitted, reused, grouped, or left unresolved.

Live One-Off Agent Variant

Use the live-agent version when you deliberately want fresh editorial judgment beyond the deterministic lexical matcher—for example, when differently worded takes express the same idea. Unlike the offline Library workflow, this route requires Automation Agent through MCP, a configured agent, model processing time, and any token costs of the selected provider.

Using Automation Agent in Premiere Pro, create a rough-select comparison sequence from the currently selected source clips in the Project panel.

Create the new sequence inside the Project bin named `Automation Agent Output`. If that bin does not exist or is not writable, stop and ask me to create or approve the output bin before continuing.

Work transcript-first:
1. Check whether every selected clip has a usable transcript.
2. If any transcript is not ready, use the safest available workflow to make it available, then export and verify all transcripts before continuing.

Then analyze the spoken content and group repeated or near-duplicate deliveries of the same intended line/sentence into take groups. Treat every selected source-file boundary as a hard take boundary.

Build a new sequence that reorganizes the selected clips into these groups in script order, with these rules:
- remove empty sections, waiting time, and obvious dead air
- keep a small natural handle before and after each spoken line
- for each line group, place all viable duplicate takes at the same timeline position, stacked vertically on separate video tracks
- order takes by recording time, placing the latest take on the lowest track and earlier alternatives above it
- leave a small fixed gap between groups so the sequence is easy to review
- preserve all source clips non-destructively

Goal: deliver a pre-sorted review timeline where each intended line appears once as a grouped stack of alternate takes, with duplicate takes aligned vertically and unusable gaps already removed.

If grouping is ambiguous, log the ambiguity instead of guessing.

Expected Result In Premiere

The result should be a new, non-destructive review sequence where:

• each intended line appears once as a group
• duplicate takes for that line are aligned vertically
• fixed gaps between groups make navigation simpler
• the original source clips remain untouched

This is meant as a review timeline, not as a final locked edit.

Variants

You can adapt this workflow depending on how you want the agent to interpret the source material.

1. Work from a selected clip in a timeline instead of the Project panel

If your source is already selected in an active sequence, change the prompt so it works from the currently selected timeline clip rather than from the selected project item.

That is useful when:

• you already marked or trimmed the clip in a sequence
• the source clip is easier to identify in the timeline than in the Project panel
• you want the workflow to start from an editorial context that already exists

2. Be more conservative or more generous with similar takes

The current prompt is relatively conservative. If two spoken lines are similar but not clearly the same intended line, the agent should usually avoid stacking them together and should log the ambiguity.

If that is what you want, keep the prompt as-is.

If you want more aggressive grouping, add an instruction that the agent may group slightly different deliveries together when they are clearly alternate takes of the same intended line, even if the wording is not identical.

That can be useful when:

• speakers paraphrase slightly between takes
• lines contain restarts or filler words
• you care more about review speed than strict transcript matching