Mixing & loudness-compliant delivery
Mix stems to target loudness and deliver to spec.
20 project-authored proposals · taxonomy v1.1 · 0 new benchmark runs. Verifiers and partial-credit rules below are designs, not executed results.
- Typical applications
- Pro Tools · Logic Pro · Fairlight · REAPER
- Class GUI-need label
G2· GUI-essential
Digital-employee context
Project class assignments, not demonstrated employee capability or imported benchmark coverage. Acceptance criteria and value metrics are proposals; no ROI has been measured.
Related roles
- Creative production and revision · Link by the artifact changed: vector paths, photo adjustments, composites, cuts, captions, models, repaired dialogue and loudness-compliant mixes.
Task proposals · 20
Expand to inspect the complete original wording and proposed checks. Original proposal text remains English in both interfaces.
R06.07.01.02-T01 Mix D/M/E stems to EBU R128 with music ducking
Original instruction
In DaVinci Resolve Fairlight open fixtures/doc_reel1.drp (dialogue, music, effects stems, 6 min, synthetic voices). Mix to stereo at -23 LUFS +/- 0.5 integrated, true peak <= -1 dBTP, music ducked >= 10 dB under dialogue via automation or sidechain. Deliver output/reel1_mix.wav (48 kHz/24-bit).
- Platforms
mac· macOSwin· Windows- Applications
- DaVinci Resolve (Fairlight)
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
drag· Drag & dropav· Audio/video over timevis· Visual judgment- Work product
WP05· Time-based media- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Bounce measured (ITU-R BS.1770-4): integrated -23 +/- 0.5 LUFS, true peak <= -1.0 dBTP. Music contribution estimated by stem regression: >= 10 dB lower in GT dialogue regions than in gaps. Resolve's Console/scripting API is disabled and the VM has no terminal, ffmpeg or sox; project DB shows fader/automation or sidechain in Fairlight; CLI renders lacking project state fail.
Proposed partial credit
35% integrated loudness; 20% true peak; 35% ducking; 10% format.
R06.07.01.02-T02 Export D, M and E stems that null against the mix
Original instruction
From the finished mix in fixtures/spot_mix.ptx export dialogue, music and effects stems (each through the same master processing via bus routing) so that D+M+E sums to the printed mix. Deliver output/stems/spot_DX.wav, spot_MX.wav, spot_FX.wav plus output/spot_mix.wav.
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · Nuendo
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clickdrag· Drag & drop- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro app- Proposed difficulty
- 3 / 5
Proposed verifier
Sum of the three stems minus mix: residual <= -50 dB relative to mix RMS; each stem contains only its category (cross-correlation with GT category sources); lengths and start times identical; 48 kHz/24-bit. Pro Tools' PTSL port and Nuendo's MIDI Remote scripting are disabled, the VM has no terminal, ffmpeg or sox, and WAVs not written by the DAW are logged and fail.
Proposed partial credit
50% null test; 30% stem separation correct; 20% alignment and format.
R06.07.01.02-T03 Automate music swells on three emotional beats
Original instruction
In fixtures/short_film.RPP raise the music bed by +6 dB (+/- 1) at the three GT emotional beats in notes/beats.txt, ramping over 1 s and returning over 2 s, using volume envelopes drawn on the music track. Dialogue level must not change. Render output/short_film_mix.wav.
- Platforms
mac· macOSwin· Windows- Applications
- REAPER
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
path· Continuous pathav· Audio/video over timedrag· Drag & drop- Work product
WP05· Time-based media- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
.RPP parsed: music volume envelope has points around each beat; render analysis: music stem gain +6 +/- 1 dB at each beat peak, ramp-up 0.8-1.2 s and decay 1.6-2.4 s; dialogue stem gain unchanged within 0.3 dB. ReaScript disabled and .RPP edits from outside REAPER are logged; envelope points created in session.
Proposed partial credit
45% three swells correct; 25% ramp shapes; 20% dialogue untouched; 10% render.
R06.07.01.02-T04 Deliver a 5.1 streaming mix to dialogue-gated spec
Original instruction
Mix fixtures/feature_reel3.ptx (5.1, dialogue, music, effects) for a streaming spec: -27 LKFS +/- 2 dialogue-gated, true peak <= -2 dBTP, LFE band-limited to 120 Hz, dialogue in center channel only. Deliver output/reel3_51.wav (interleaved, SMPTE order L R C LFE Ls Rs).
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · Nuendo
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
drag· Drag & dropav· Audio/video over timevis· Visual judgment- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro applong· Long-horizon- Proposed difficulty
- 4 / 5
Proposed verifier
Dialogue-gated loudness (Dolby Dialogue Intelligence-equivalent) -27 +/- 2 LKFS; true peak <= -2 dBTP all channels; LFE energy above 120 Hz <= -40 dB re LFE; dialogue stem regression: >= 95% of dialogue energy in C; channel order verified by test tones at head. Pro Tools' PTSL port and Nuendo's MIDI Remote scripting are disabled; the VM has no terminal, ffmpeg or sox.
Proposed partial credit
35% dialogue-gated loudness; 20% true peak; 25% LFE (15 pts) and channel order (10 pts); 20% dialogue in center.
R06.07.01.02-T05 Deliver a podcast at platform loudness
Original instruction
Master fixtures/podcast_ep3_mix.wav (stereo) to -16 LUFS integrated +/- 1, true peak <= -1 dBTP using Logic Pro's loudness meter and a limiter, and export output/ep3.wav (48 kHz/24-bit) and output/ep3.mp3 (128 kbps CBR, 44.1 kHz) with title and episode ID3 tags.
- Platforms
mac· macOS- Applications
- Logic Pro
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
click· Point & clicktype· Text & forms- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro app- Proposed difficulty
- 2 / 5
Proposed verifier
Both files: -16 +/- 1 LUFS, true peak <= -1.0 dBTP (MP3 decoded); MP3 bitrate/sample rate per spec; ID3 TIT2 and TRCK match notes; limiter not causing > 3 dB gain reduction 99th percentile. Control pair is T11. AppleScript/System Events UI scripting is blocked for the agent and the VM has no terminal, ffmpeg, sox or LAME; files not written by Logic are logged and fail.
Proposed partial credit
40% loudness; 25% true peak; 20% formats; 15% tags.
R06.07.01.02-T06 Prepare a 30-second commercial to broadcast spec
Original instruction
Finish fixtures/spot30.ptx for US broadcast: ATSC A/85 -24 LKFS +/- 2, true peak <= -2 dBTP, program exactly 00:00:30:00 at 29.97 DF starting 01:00:00:00, 2-pop at 00:59:58:00, BWF with timecode stamp and description metadata. Deliver output/spot30_A85.wav.
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · DaVinci Resolve (Fairlight)
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clickdrag· Drag & dropav· Audio/video over time- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro app- Proposed difficulty
- 3 / 5
Proposed verifier
BWF bext parsed: TimeReference = 01:00:00:00 minus pre-roll per spec, description present. Loudness -24 +/- 2 LKFS, true peak <= -2 dBTP; 2-pop 1 frame of 1 kHz exactly 2 s before program; program length 30 s +/- 0 frames (silence detection). Pro Tools' PTSL port and Resolve's Console/scripting API are disabled, the VM has no terminal, ffmpeg, sox or BWF tools, and outside WAV writes fail.
Proposed partial credit
30% loudness; 15% true peak; 30% timing (2-pop, length, TC); 25% BWF metadata.
R06.07.01.02-T07 Fold a 5.1 mix down to stereo without dialogue loss
Original instruction
Create a stereo Lo/Ro fold-down of fixtures/trailer_51.wav in Fairlight or Pro Tools with center -3 dB, surrounds -3 dB, LFE dropped, then adjust bus compression so the stereo hits -24 LUFS +/- 1 with dialogue level within 1 dB of the 5.1 version. Deliver output/trailer_LoRo.wav.
- Platforms
mac· macOSwin· Windows- Applications
- DaVinci Resolve (Fairlight) · Pro Tools
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
click· Point & clickdrag· Drag & dropvis· Visual judgment- Work product
WP05· Time-based media- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Fold-down coefficients estimated by regression of output on input channels: C 0.707 +/- 0.03, Ls/Rs 0.707 +/- 0.03, LFE ~0; integrated -24 +/- 1 LUFS; dialogue-gated level within 1 dB of 5.1; no clipping (true peak <= -1 dBTP). Resolve's Console/scripting API and Pro Tools' PTSL port are disabled; the VM has no terminal, ffmpeg or sox, and WAVs not written by the DAW are logged.
Proposed partial credit
40% downmix coefficients; 30% loudness; 20% dialogue level; 10% peaks.
R06.07.01.02-T08 Balance dialogue against a busy effects bed
Original instruction
Scene fixtures/market_scene.RPP has crowd and traffic effects that mask dialogue. Rebalance with automation and EQ carving so dialogue-to-background ratio is >= 10 dB in every dialogue phrase while background stays >= -35 LUFS short-term in gaps (scene must still feel busy). Render output/market_mix.wav.
- Platforms
mac· macOSwin· Windows- Applications
- REAPER · Pro Tools
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
path· Continuous pathav· Audio/video over timevis· Visual judgment- Work product
WP05· Time-based media- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Stem-regression measure per GT phrase: dialogue minus background short-term loudness >= 10 LU in >= 95% of phrases; background short-term loudness in gaps >= -35 LUFS; EQ dip in 1-4 kHz on FX bus present in session; ESTOI of mix vs clean dialogue >= 0.85. ReaScript and Pro Tools' PTSL port are disabled; .RPP edits and WAVs written outside the DAW are logged and fail.
Proposed partial credit
45% per-phrase ratio; 20% busy background kept; 20% intelligibility; 15% session EQ/automation.
R06.07.01.02-T09 Reduce loudness range for mobile delivery
Original instruction
The theatrical mix fixtures/feature_stereo_lra22.wav has LRA 22 LU. Create a mobile version with LRA <= 10 LU at -16 LUFS +/- 1, true peak <= -1 dBTP, using compression and dialogue-anchored automation so quiet dialogue rises without pumping. Deliver output/feature_mobile.wav.
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · Logic Pro · DaVinci Resolve (Fairlight)
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
drag· Drag & dropav· Audio/video over timevis· Visual judgment- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro app- Proposed difficulty
- 3 / 5
Proposed verifier
EBU Tech 3342 LRA <= 10 LU; integrated -16 +/- 1 LUFS; true peak <= -1 dBTP; quietest-quartile dialogue phrases raised >= 6 dB; pumping metric (gain-reduction modulation 2-8 Hz) below threshold. Pro Tools' PTSL port, Logic's AppleScript/UI scripting and Resolve's Console/scripting API are disabled; the VM has no terminal, ffmpeg or sox, and outside WAV writes are logged.
Proposed partial credit
35% LRA; 25% loudness/peak; 25% quiet dialogue lifted; 15% no pumping.
R06.07.01.02-T10 Export a Dolby Atmos master from Logic Pro
Original instruction
Mix fixtures/atmos_promo.logicx (bed 7.1.2 plus 6 objects) so the 3 GT sound objects move along the trajectories in notes/object_paths.json, dialogue in the center bed, then export an ADM BWF master and a binaural re-render. Loudness -18 LKFS +/- 1, true peak <= -1 dBTP.
- Platforms
mac· macOS- Applications
- Logic Pro
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
drag· Drag & dropgeom· Precise geometryav· Audio/video over time- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro applong· Long-horizon- Proposed difficulty
- 5 / 5
Proposed verifier
ADM BWF parsed (ADM XML): 10-channel bed + 6 objects; object position automation matches paths (mean error <= 0.1 in normalized room coords); dialogue in C; Dolby renderer loudness -18 +/- 1 LKFS, true peak <= -1 dBTP; binaural file present and length-matched. AppleScript/System Events UI scripting is blocked and the VM has no terminal or ADM/Dolby CLI tools; files not written by Logic are logged.
Proposed partial credit
20% ADM structure; 35% object trajectories; 15% dialogue in center; 30% loudness/peaks (20 pts) and binaural (10 pts).
R06.07.01.02-T11 Normalize the podcast with ffmpeg loudnorm (control pair of T05)
Original instruction
Control pair of T05: from a terminal only, normalize fixtures/podcast_ep3_mix.wav with two-pass ffmpeg loudnorm to -16 LUFS / -1 dBTP and produce output/ep3_cli.wav and output/ep3_cli.mp3 (128 kbps CBR, 44.1 kHz) with the same ID3 tags.
- Platforms
mac· macOSwin· Windowslin· Linux- Applications
- ffmpeg
- Interfaces
term· Terminal- GUI need · task setting
CL· CLI-sufficient- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
code· Code & commands- Work product
WP05· Time-based media- Conditions
- None declared
- Proposed difficulty
- 2 / 5
Proposed verifier
Same checks as T05 on the CLI outputs: loudness, true peak, formats, tags. Scored side by side with T05 to measure the GUI-vs-CLI delta.
Proposed partial credit
40% loudness; 25% true peak; 20% formats; 15% tags.
R06.07.01.02-T12 Route a mix session into DX, MX and FX buses with stem prints
Original instruction
In fixtures/blank_session.RPP with 24 unsorted tracks (named per notes/track_list.csv), create DX, MX and FX folder buses feeding a master, route each track by category, add a stem-print track per bus recording post-fader, and set master to 48 kHz. Save output/routed.RPP.
- Platforms
mac· macOSwin· Windows- Applications
- REAPER · Pro Tools
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
drag· Drag & dropclick· Point & click- Work product
WP08· App/system state- Conditions
- None declared
- Proposed difficulty
- 2 / 5
Proposed verifier
.RPP parsed: 3 buses exist; each of 24 tracks routed only to its category bus per CSV; buses feed master; 3 print tracks receive post-fader sends from buses and are record-armed; project sample rate 48000. ReaScript and Pro Tools' PTSL port are disabled, and .RPP edits made outside REAPER are logged and fail.
Proposed partial credit
30% buses; 40% track routing correct; 20% print tracks; 10% project settings.
R06.07.01.02-T13 Ride dialogue levels for a consistent scene
Original instruction
In fixtures/sc4_dialogue.ptx, ride the dialogue fader with volume automation (no compressor on DX) so short-term loudness of dialogue phrases stays within +/- 3 LU of -24 LUFS across the scene, which currently swings 14 LU between shouting and whispering. Bounce output/sc4_dx_ridden.wav.
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · REAPER · Logic Pro
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
path· Continuous pathav· Audio/video over timedrag· Drag & drop- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro app- Proposed difficulty
- 4 / 5
Proposed verifier
Per GT phrase short-term loudness within -27 to -21 LUFS for >= 90% of phrases; session has volume automation breakpoints on DX and no dynamics plugin; automation slopes avoid jumps > 6 dB within 50 ms; no clipping. Pro Tools' PTSL port, ReaScript and Logic's AppleScript/UI scripting are disabled; the VM has no terminal, ffmpeg or sox, and outside WAV writes are logged.
Proposed partial credit
55% phrase consistency; 25% automation-only method; 20% smooth rides.
R06.07.01.02-T14 Build an M&E track for international versions
Original instruction
From fixtures/ep4_mix.ptx create an M&E stereo track: remove dialogue but keep production effects (footsteps, doors) by pulling PFX from the dialogue tracks into an FX track and filling holes with Foley from fixtures/foley_lib/, so effects are continuous where dialogue was removed. Deliver output/ep4_ME.wav.
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · Nuendo
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
drag· Drag & dropav· Audio/video over timeclick· Point & click- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro applong· Long-horizon- Proposed difficulty
- 4 / 5
Proposed verifier
Speech detector on output: speech presence <= 1% of frames; PFX events from GT list present in >= 85% (onset within 2 frames); no holes (gaps where full mix had effects) longer than 200 ms; loudness within 2 LU of the full mix's M+E portion. Pro Tools' PTSL port and Nuendo's MIDI Remote scripting are disabled; the VM has no terminal, ffmpeg or sox, and outside WAV writes are logged.
Proposed partial credit
35% dialogue removed; 35% PFX retained/filled; 20% continuity; 10% loudness.
R06.07.01.02-T15 Pan effects to on-screen positions in a 5.1 mix
Original instruction
Pan the 14 spot effects in fixtures/chase_51.ptx to match on-screen source positions (GT screen x per frame in notes/fx_positions.csv): left-right across L/C/R and off-screen pass-bys into surrounds, with automation following moving sources. Deliver output/chase_51.wav.
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · DaVinci Resolve (Fairlight)
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
drag· Drag & dropav· Audio/video over timegeom· Precise geometry- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro app- Proposed difficulty
- 4 / 5
Proposed verifier
Per effect, energy-weighted panning angle from channel ratios vs GT screen x: mean error <= 15 deg for on-screen; off-screen pass-bys have surround energy >= 50% at GT frames; moving sources' pan trajectories correlate >= 0.8 with GT. Pro Tools' PTSL port and Resolve's Console/scripting API are disabled; the VM has no terminal, ffmpeg or sox, and WAVs not written by the DAW are logged.
Proposed partial credit
40% static pan accuracy; 35% moving trajectories; 25% surround pass-bys.
R06.07.01.02-T16 Assemble a full deliverables package for a short film
Original instruction
Per notes/deliverables_spec.pdf, deliver from fixtures/shortfilm_final.ptx: stereo printmaster (-24 LKFS), 5.1 printmaster (-27 LKFS dialogue-gated), M&E stereo and 5.1, DX/MX/FX stems for both, with 2-pops, BWF timecode, and the exact folder/file naming scheme. Write everything under output/deliverables/.
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · Nuendo
- Interfaces
desk· Desktop GUI- GUI need · task setting
HY· Hybrid- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clickdrag· Drag & dropav· Audio/video over time- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro applong· Long-horizon- Proposed difficulty
- 5 / 5
Proposed verifier
Folder tree and names match spec exactly; each file's channel count, loudness and true peak per spec; stems null against their printmaster (<= -50 dB); M&E speech presence <= 1%; 2-pops present and aligned; BWF TimeReference consistent across files.
Proposed partial credit
15% structure/names; 30% loudness/peaks per file; 25% stem nulls; 30% M&E (15 pts) and sync metadata (15 pts).
R06.07.01.02-T17 Apply a mid-session client mix revision
Original instruction
Start mixing fixtures/promo60.RPP to spec (-23 LUFS). A client note arrives in notes/revision_2.txt during the task (music -4 dB in 0:20-0:35, VO line at 0:42 louder by 3 dB, add a whoosh from fixtures/whoosh.wav at 0:51). Apply all notes without breaking spec and render output/promo60_v2.wav.
- Platforms
mac· macOSwin· Windows- Applications
- REAPER · Pro Tools
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
drag· Drag & dropav· Audio/video over timeclick· Point & click- Work product
WP05· Time-based media- Conditions
interrupt· Interruptions- Proposed difficulty
- 4 / 5
Proposed verifier
Stem regression: music -4 +/- 1 dB in 0:20-0:35 only; VO line at 0:42 +3 +/- 1 dB; whoosh onset within 2 frames of 0:51; integrated -23 +/- 0.5 LUFS, true peak <= -1 dBTP; other regions within 0.5 dB of v1 mix where unaffected. ReaScript and Pro Tools' PTSL port are disabled; .RPP edits and renders written outside the DAW are logged and fail.
Proposed partial credit
20% music change; 20% VO change; 20% whoosh; 40% spec held (25 pts) and rest unchanged (15 pts).
R06.07.01.02-T18 Line up a 2-pop and tail pop with picture
Original instruction
In fixtures/reel5_sync.drp, the mix is 3 frames late against picture. Using the picture's 2-pop flash frame and tail pop frame, slip the whole mix so the 2-pop and tail pop land exactly on their flash frames at 23.976 fps, without trimming program audio. Render output/reel5_synced.wav.
- Platforms
mac· macOSwin· Windows- Applications
- DaVinci Resolve (Fairlight) · Pro Tools
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
CL· CLI-sufficient- Interaction primitives
drag· Drag & dropav· Audio/video over timevis· Visual judgment- Work product
WP05· Time-based media- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Rendered audio: 2-pop onset within the GT flash frame (+/- 0 frames) and tail pop within its frame; program cross-correlation with original shows an offset of exactly -3 frames; no content trimmed (length equals original). Resolve's Console/scripting API and Pro Tools' PTSL port are disabled; the VM has no terminal, ffmpeg or sox, and WAVs not written by the DAW are logged.
Proposed partial credit
40% head sync; 30% tail sync; 30% program intact.
R06.07.01.02-T19 Build a master bus chain to hit loudness with dynamics intact
Original instruction
On the master of fixtures/trailer_mix.drp build a chain (bus compressor 2:1 slow attack, EQ, true-peak limiter) so the trailer hits -14 LUFS for web, true peak <= -1 dBTP, while PLR (peak-to-loudness ratio) stays >= 9 dB and no limiter gain reduction exceeds 4 dB. Render output/trailer_web.wav.
- Platforms
mac· macOSwin· Windows- Applications
- DaVinci Resolve (Fairlight) · Logic Pro
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clickdrag· Drag & dropvis· Visual judgment- Work product
WP05· Time-based media- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Integrated -14 +/- 0.5 LUFS; true peak <= -1 dBTP; PLR >= 9 dB; estimated limiter gain reduction (output vs pre-limiter tap) max <= 4 dB; chain present in project (plugin params read). Resolve's Console/scripting API and Logic's AppleScript/UI scripting are disabled; the VM has no terminal, ffmpeg or sox, and WAVs not written by the DAW are logged.
Proposed partial credit
35% loudness; 20% true peak; 30% dynamics preserved; 15% chain present.
R06.07.01.02-T20 Mix a 22-minute TV episode to broadcaster spec
Original instruction
Mix fixtures/ep101_premix.ptx (dialogue, music, effects premixes, 22 min, synthetic voices) to notes/broadcaster_spec.pdf: stereo -23 LUFS +/- 0.5, true peak -1 dBTP, LRA <= 15 LU, dialogue always intelligible over music, plus clean M&E. Deliver output/ep101_mix.wav and output/ep101_ME.wav.
- Platforms
mac· macOSwin· Windows- Applications
- Pro Tools · DaVinci Resolve (Fairlight) · Nuendo
- Interfaces
desk· Desktop GUI- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
path· Continuous pathav· Audio/video over timedrag· Drag & drop- Work product
WP05· Time-based media- Conditions
licensed· Licensed pro applong· Long-horizon- Proposed difficulty
- 5 / 5
Proposed verifier
Loudness, true peak and LRA per spec; dialogue-to-music ratio >= 8 LU in >= 95% of dialogue phrases (stem regression); ESTOI vs clean dialogue >= 0.85; M&E speech presence <= 1% and loudness consistent; session automation present on DX/MX. Pro Tools' PTSL port, Resolve's Console/scripting API and Nuendo's MIDI Remote scripting are disabled; the VM has no terminal, ffmpeg or sox.
Proposed partial credit
30% loudness/peak/LRA; 30% dialogue clarity; 25% M&E; 15% session evidence.
Mapped research references
No benchmark reference is assigned to this class in the current mapping.