FIELD GUIDE · SOUND
Sound and music:
made against the cut.
Sound is made in Edit, against the cut. A music cue covers a range of the timeline and is a row of sections that turn only where the picture does; effects and ambience are made from a marked range and land on their tracks. Describe the sound the way a director talks to a composer; the price shows before anything is made.
07b / SOUND
Score the picture: music that turns on the cuts.
Music: a cue in detail
What it is. Since 0.18 the music is made in Edit, against the cut (see Sound in Edit under Edit); there is no Sound stage and no Score these shots on the Shot List. A cue is one piece of music: made in Edit, it covers a range of the cut; a cue made on the 0.17 Sound tab covers a run of shots and stays readable, on the music track. Open the cue in the sound library opens it in detail, as below.
Start a cue. In Edit, mark a range and press Make sound for this range with Music chosen; or ask the Edit chat ("a lonely piano theme under scene 2"). The cue is named after the place ("PINE-WOODS"); rename it any time.
Where the timing comes from. A cue made in Edit is timed by its range: the shots under it, each for the seconds it plays inside the range. If the cut under the range changes, the cue says so and Start again from the cut drafts its sections again. A 0.17 cue takes its shots' lengths from the cut when they play there as one run, else the Shot List's.
Sections turn on the cuts. The music changes only where the picture does, so a cue is a row of sections, each covering whole shots. The first draft is free and comes from the shots' own words (their why, what happens in them, the beats they cover). Each section has:
- a device: what the harmony does there (below);
- the picture it plays under, in your words ("she steps out from behind the tree"). This is for you and for Draft the score; the music model never reads it;
- more music words for that section only ("solo piano enters, quieter").
Merge with next removes a turn; Turn at 1C adds one on that cut. A section needs at least 3 seconds; a shorter one is marked and must be merged before you generate.
The devices. Ten ways for the harmony to serve a picture:
- Theme: the cue's theme, stated plainly.
- Dread: a held minor chord with one voice creeping up a half step and back. Suspense, danger, something wrong.
- Reveal: the minor chord opens into its major and the line climbs. Someone appears; a door opens.
- Loneliness: a jump to a distant major key a third away, with the bright raised fourth of the Lydian mode. A person alone; wonder.
- Longing: up a half step into a new key, chords and melody rising together. Hope, rising stakes.
- Release: an unexpected resolution home. Relief, arrival.
- Melodrama: plain in-key chords over a held bass note, old-fashioned and warm. Romance, memory.
- Pulse: a steady ostinato. Motion, urgency, a chase.
- Hush: the music thins to one held note, or nearly nothing. A held breath.
- Return: the opening material comes back, changed. An ending that answers the beginning.
What the director says. Describe the cue the way a director talks to a composer: where we are, what we see, what someone feels. No chords needed. Draft the score reads your words with the shots and the scenes' mood and beats, and proposes a theme (key, tempo, instruments, character) and the sections with their devices. It shows its price first (Draft the score: get the cost, then Draft · up to N credits) and arrives as a card: Use this draft replaces the sections and the theme, Ignore keeps yours. Style words that name a work or a person are left out and listed on the card. Start again from the cut (a 0.17 cue: from the shots) makes the free draft again and keeps your words and theme. In the Make sound pop-up, only the musical words of what you want become the theme (instruments, a key, a tempo, the character); the picture words stay as the director's words.
The theme. Key, tempo, instruments and the cue's character hold the whole cue together. Leave them empty and the music chooses; a small chamber ensemble is the default.
Sent exactly as shown. Under the sections is the plan the music model receives (ElevenLabs Music v2.5, instrumental): one section per row with its time and its music words, the theme in the first row, and what is never wanted (vocals, lyrics). Section lengths are exact, so the music turns on the cuts.
Generate. Choose Credits or Your ElevenLabs key. On credits, Estimate shows the most one take and three takes can cost; each take settles on its measured length and the rest is returned. Generate 1 or Generate 3 saves the project first, then the takes are made on the server (they land in a minute or two, named like takes: S01_MUS_PINE-WOODS_TK02). A refused plan releases its credits and says why.
Love and Like. The first take of a cue made in Edit lands on the music track at its range by itself, marked TEMP. Love puts a take on the music track (at the range, or where a 0.17 cue's shots start), at 70% level with short fades; the next take you love replaces it in the same place, keeping any level or position you changed. Un-loving the placed take takes it off the track. Like keeps a take in the library. Mark final removes the TEMP mark from the cue and its clips; Remove cue takes it and its clips away (the files stay in the project's media).
When the cut changes. Each take remembers the section lengths it was made for, and the clip on the music track remembers the picture under it. After a re-cut, the take says what moved ("The cut changed since this take: 1C is 1 s shorter.") and the clip says it no longer fits the cut; make a new take to fit, or keep it with It fits.
Known limits. Sections turn on cuts, not yet on a beat inside a long shot. The music track is mixed at the level you set; the music does not yet dip under dialogue on its own, and does not follow the picture when the cut changes. One live call against ElevenLabs is still owed.
What changed in 0.20 (round 8): music the script calls for goes to Edit's Sound tab
- Check script catches it. Under the findings, Music the script calls for lists every cue the script asks for: a song or singing, humming, a choir, a band, a radio or a record playing, an instrument played, music swells. Each has its scene, the lines where it starts and ends (the line that stops it, else the scene's last line), and its kind. It is found in code, the same way every time; a fridge that hums, a kettle that sings or a radio that reads the news is not music. It is a list, not a finding to fix.
- Performed on screen, source or score. Someone in frame who sings or plays is performed on screen: that is a sync part of the take, so the take's model must have sound. The cue made in Edit is then the bed or the accompaniment under it; it never replaces the performance. Music heard in the scene (a radio, a record, singing from another room) is source music; music the script asks for over the scene is score.
- Lock sends it to Sound. In Edit, Music the script calls for shows each cue with the script's words (shown for you, never sent to the music model), its kind, and the range of the cut it covers: a performed cue covers the clips of its moment (the shots whose own words name the performance), any other its scene. A scene not in the cut yet says so.
- Make for this range opens Make sound for this range on that range, with the brief empty: you write what you want, as in 0.18. The first take lands on the music track at the range, as that music. Placing a take of it by hand does the same. Dismiss sets aside music the film does not need; Bring back undoes it.
- Before export. Export lists the scripted cues not on the music track ("2 scripted music cues not in the cut"). The export can still go ahead; it says so. For the agent:
GET /api/projects/:id/music/scriptedlists each cue with its range and status, and the export's note.
Effects and ambience made in Edit (0.22)
What it is. On the Sound effects and Ambience tracks, Make sound for this range now makes the sound, with ElevenLabs Sound Effects (the same ElevenLabs account and key as the music: Credits, or Your ElevenLabs key). A new project needs nothing to import.
Mark a range, describe the sound. Mark in and out on the timeline, press Make sound for this range, choose Sound effects or Ambience and say what makes the sound, how, where and how near ("a heavy wooden door slams in a stone stairwell", "steady rain on a canvas awning, distant traffic"). The result lands on that track at the range.
- An effect fits its moment. Its length is the range's, from 0.1 to 30 seconds. The model makes at least half a second, so a shorter moment is made at 0.5 s and placed cut to the moment, ending on a short fade.
- Ambience fills the range. Up to 30 seconds it is made at the range's length, as a loop. A longer range (a whole scene) is made as one loop of an equal share of the range (a 47 s range: one 23.5 s loop, twice) and repeated in the studio, sample for sample, to the exact length. The repetition is free: you pay for one generation. Before repeating it, the studio trims any silence the file's encoding left at its ends, measures the seam (the jump from its last sample into its first, and the level across it) and, only when the seam is not clean, blends the end into the start over 50 milliseconds. The job keeps the measure.
Only the sound's own words are sent. Sent exactly as shown has the text the sound model reads. People's names, quoted lines, shot labels ("2B"), scene numbers, times and camera words ("close-up", "the camera pans across") are left out and listed beside it. Ambience is sent as a steady background bed, with no music. The picture's own words (the shot's description, the scene, the script) never go to the sound model.
The price first. The price line shows what a version is expected to cost and the most it can cost, in credits, before anything is made. Each version reserves its ceiling, settles on the seconds made, and the rest is returned; a refused or failed call releases its credits, and an unknown outcome (a timeout, no answer) is held for review. Make 1 or Make 3 versions; on your own key nothing is charged in credits. Import a file is still there for your own recordings.
Every result is a version in the library. Named like the takes (S02_AMB_RAIN-ON-THE-AWNING_V01, S01_FX_A-HEAVY-WOODEN-DOOR_V02), by kind, scene and version. The first version of a batch is placed on its track at the range; the others wait in the library to be placed or dragged. An ambience made from a loop says so ("loop 23.5 s × 2"). While they are made, Being made lists them; one that failed says why.
Sounds the script names. Lock reads the locked draft for the sounds it names, in code and the same way every time, and Edit lists them under Sounds the script names, with the script's sentence (shown for you, never sent), effect or ambience, and where each stands (to make, made, on its track):
- a sound word in capitals ("The door SLAMS");
- something that sounds, doing what makes the sound: a door slams or creaks, a phone or a bell rings, glass shatters, a dog barks, an engine roars, a kettle whistles, a clock ticks, rain drums on a roof, wind howls, waves crash, a crowd murmurs;
- a sound named as heard ("we hear the hum of a generator", "the sound of …");
- the weather the scene heading names ("- RAIN").
Continuing sounds (weather, water, traffic, crowds, insects, birdsong, a hum, a ticking clock, a far-off dog) are ambience and cover their scene; the rest are effects at their moment: the clip whose shot's words (or its timeline) name what makes the sound, for two seconds that you can move. Speech is never read, music is the music list's, and a sound denied or ending ("no wind", "the rain stops", "the name rings a bell") is not listed. Make for this range opens the pop-up on that range with the sound's own words as the description, price first; Dismiss sets one aside. A project locked before 0.22 has Find them now · free (or lock again). For the agent: GET /api/projects/:id/sfx/scripted; POST /api/projects/:id/sfx/quote and POST /api/projects/:id/sfx (kind, start, end, text, takes, payment, maxCredits, scriptSoundId).
They play in the render like any sound on the tracks: the effects track at its level, ambience lower, with a second's fade at each end.
Known limits. Three versions at most per click. Ambience fills at most ten minutes at a time: mark a longer stretch in parts. The loop seam is measured and blended in code; it has not had a listening review by a person. The price uses the published API rate ($0.12 a generated minute) doubled as the ceiling until a live bill confirms it.
What changed in 0.22 (TAR-263): performed music by notes and voices
Music the script calls for also catches performed music written without "sing" or an instrument played: a note sung, held or hit ("holds the top note", "hits the note"), a performance caught in progress ("breaks off mid-phrase", "halfway through the hymn"), which ends on that line, voices joined in harmony or in unison, a singer named by their voice (tenor, baritone, contralto, mezzo, countertenor, vocalist) and a piece performed ("rehearses the aria"). A written note, a banknote, "a note of doubt" or a note held out in a hand is not music.
What changed in 0.22 (§22): placed sounds play in Edit's preview
Edit's Program viewer now plays the sound tracks with the picture: music, effects, ambience, narration and a take's sound unlinked to Dialogue, each at its range, its level and its fades, the way the render mixes them. A muted track is silent in the preview and left out of the render alike. Before, the preview played only the visible take's own sound, so a placed cue sounded only in the render. The viewer plays forward at the shuttle's speed; stopped, scrubbing or shuttling back it plays no placed sound. A level above 100% plays at full in the preview (the render keeps it).
QUESTIONS · SOUND
Questions, answered from the guide.
- How do I add music to the cut?
Start a cue. In Edit, mark a range and press Make sound for this range with Music chosen; or ask the Edit chat ("a lonely piano theme under scene 2"). The cue is named after the place ("PINE-WOODS"); rename it any time.
- How do I make a sound effect or ambience?
Mark a range, describe the sound. Mark in and out on the timeline, press Make sound for this range, choose Sound effects or Ambience and say what makes the sound, how, where and how near ("a heavy wooden door slams in a stone stairwell", "steady rain on a canvas awning, distant traffic"). The result lands on that track at the range.
- What happens to a music cue when I re-cut?
When the cut changes. Each take remembers the section lengths it was made for, and the clip on the music track remembers the picture under it. After a re-cut, the take says what moved ("The cut changed since this take: 1C is 1 s shorter.") and the clip says it no longer fits the cut; make a new take to fit, or keep it with It fits.
- What's the difference between performed music, source music and score?
Performed on screen, source or score. Someone in frame who sings or plays is performed on screen: that is a sync part of the take, so the take's model must have sound. The cue made in Edit is then the bed or the accompaniment under it; it never replaces the performance. Music heard in the scene (a radio, a record, singing from another room) is source music; music the script asks for over the scene is score.