Your cast already tells you how many mics the room needs — you just counted it
A multi-voice cast decision is not taste. It is principals plus largest share. Principals are speaking roles with more than 5% of dialogue. Largest share is the share of the biggest role. The two numbers together answer whether a single skilled narrator with character voices carries the cast, whether a duet with two leads sharing the mic pays, or whether an ensemble needs multi-voice where each principal has a voice the listener can track without guessing who said what.
Three leads at 45% largest share is single. A skilled single does character voices for three principals where the lead at 45 still carries, because the listener can hold three voices and the lead's 45 is enough to anchor chapters. The tool reports single — character voices carry the cast — not because multi cannot be bought but because single is tighter and cheaper and the extra $900 per extra voice plus base that multi adds over single mid $2,430 at 8.1 hrs for 75k is overhead the story does not buy clarity with.
Six leads at 35% largest share is multi. No single lead carries where the largest at 35 is less than a third and five other voices share the rest. The ensemble threshold the tool uses — four or more principals at largest below 40% — is the line where a single narrator must invent voices for a cast the listener cannot hold as one timbre. The tool reports multi-voice — ensemble pays — not because single cannot be performed but because multi adds orientation the listener feels as who said what without a narrator tag. Fantasy ensemble needs multi sooner than thriller lead-heavy where a single voice can carry tight dialogue, but at six principals the threshold is passed regardless of genre.
Two leads at 50% each is duet. Two principals where each carries about half the dialogue is the duet window at 35 to 60% largest share for a cast of two. The duet is two voices sharing, not full cast. Romance at two leads 50 is duet the same way thriller at two leads 50 is duet: the story is two leads, each needs a voice the listener can track as lead versus lead.
One principal is solo by definition. Single is the only sensible even when the single is at 90% because there is no second voice to cast.
Large cast but lead heavy is the warning case. Four or more principals but largest at 60% or more is contradictory: the cast suggests ensemble but the share says lead heavy. The lead still carries; multi would be under-used because the listener would rarely hear the ensemble after the lead's 60 dominates. The tool reports warning — large cast but lead heavy — not because the numbers are forbidden but because the shares suggest the book's dialogue is lead heavy despite the cast list. The writer should re-count largest share or re-count principals; often the four includes cameos below 5% that should not be counted as principals.
Where the numbers come from, shape by shape
The breakdown reports why this recommendation. Cast size is principals the listener must track, not cameos. Largest share is lead heavy at 60 plus is single, ensemble below 40 is multi. Single cost baseline is 8.1 hrs mid for 75k at $2,430; scales with hours at the rate the hour count from the production calculator implies. Multi extra is casting plus session overhead at about $900 per extra voice plus base at $1,200. Genre is the note that ensemble needs multi in fantasy more than thriller. The metrics report cast size, largest share, recommendation (single, duet, multi), and extra cost over single mid.
The table shows the same cast logic at a glance: cast 1 any single solo, cast 2 at 35 to 60% duet, cast 3 at 60 plus single lead heavy, cast 4+ below 40% multi ensemble, 5 plus any multi large ensemble. The table is not a menu of cast sizes to choose from to land an easier recommendation; the cast and share are what the book has.
Reading the genre shift with the same words
The voice casting sheet already maps genre to pitch and pace: thriller low-mid tight at 150, fantasy mid warm balanced at 145, romance mid-warm slow at 140, horror low still slow at 138. The cast planner uses genre the same way as a note that ensemble needs multi in fantasy more than thriller. A fantasy with six leads at 35% is multi the same way a thriller with three leads at 45% is single, but the threshold at four principals below 40% is the line where fantasy crosses sooner. The 40 total at audio cost and the 41 at ACX are siblings from the same hours; cast plan is principals and share from dialogue, not word count.
Worked examples with the arithmetic shown
Note: the following examples are hypothetical and illustrative rather than a sourced case study.
A thriller three leads at 45% — single. Cast size 3 principals at largest share 45. The cast is not ensemble because largest below 40 requires four plus. The largest at 45 is not lead heavy at 60, so the middle: single where character voices carry. The cost baseline single mid $2,430 at 8.1 hrs, multi extra about $3,900 for cast of three at $900 per extra voice plus base would be wasted clarity. The verdict is pass — single. The metrics show Recommendation single.
A fantasy six leads at 35% — multi. Cast size 6 principals where largest share 35 is below 40 and ensemble threshold at four plus below 40 is met. Multi-voice adds clarity single cannot with character voices alone. The tool reports multi-voice ensemble pays. The metrics show Recommendation multi.
Two leads at 50% — duet. Cast size 2 principals at 50% each is the duet window where two leads share 35 to 60% each. The story is two leads, each needs a voice. The metrics show Duet.
Why the tool stops at recommendation, not booking
The recommendation is the cast count and share the room can repeat: genre threshold and cost over single mid. The booking that follows is which narrator reads which principal, but the recommendation that precedes it is whether the room needs one mic or many.
Common mistakes and how to avoid them
Counting cameos as principals is the first. A speaking role with more than 5% of dialogue is a principal; a cameo at 2% should not be counted. Re-count largest share correctly.
Using a single where multi pays is the second. At four plus principals below 40% the listener loses orientation with single.
Using multi where duet pays is the third. Two leads at 50% is duet, not full cast.
Forgetting to budget extra is the fourth. Multi extra about $900 per extra voice plus base over single mid $2,430 at 8.1 hrs for 75k.
Estimating cast from pages is the fifth. Cast plan is principals and share, not word count. A 75,000-word thriller that Runtime Estimator says is 90 minutes and Scene Counter says is 88 scenes has its cast as principals and share, not hours.
Final checks before you commit and upload for audio
Run the tool twice with two tier splits — one as single versus multi for the same cast and share. The estimate should bracket the true cast plan within a one-cast window.
Read the breakdown with the genre note.
Compare the scope to the cover. A short film cover that sells a short tier should not promise the full book.
Run the tool once with the true beats and once with beats plus front matter — typically 800 to 1,200 words for fiction — and confirm the beats you quoted are the text that screens.
A note on very short manuscripts that expand for audio
A 20,000-word manuscript at 950 blended is 21 scenes. The cast at four parts can still be built.
Why this page sits where it does in the audio cluster
Audiobooks needs ten numbers before any blade touches outline: hours, cost, ACX specs, casting, and now cast plan. This page adds multi-voice.
Engineering note — the table is the map for cast plan
The table that shows cast size and share → recommendation is the edit decision list. It is not a menu.
Book to audio heptad plus cast — hours, cost, specs, casting, and now multi-voice
Book to audio needs seven numbers before any mic is booked: hours, cost, ACX specs, voice casting, and now multi-voice. Hours from the finished hour estimator tells the studio the length; cost from the PFH calculator tells the budget; ACX specs from the checker tells whether the file passes the door; voice casting from the sheet tells whose voice fits; multi-voice from this planner tells how many mics the room needs. The seven are the deck the producer can hand to the sound supervisor without re-explaining the book.
Engineering note — the divisor chain with cast plan
Words per page is 275 at 6×9 Garamond 12 on 15.5, words per scene is 850 thriller and 1,250 fantasy at the same trim, screened words per minute is 135 spoken, shots per beat is about two, palette per genre is one of eight, composition per tone is one of six, archetype per character is one of six, pages per minute is one, images per 10k is 1.2 to 6.0, trailer beats per length is 3 or 5 or 6, audio hours per words is 9,300 at 150 wpm, cast principals at 1.2 to 6.0 is the cast the audio needs. The chain's input is the same word count and the four parts you named plus hook and stakes. The outputs are the twelve masters plus the schedule that cast drives.
A note on very episodic manuscripts above 180 scenes again in audio cast terms
A 200,000-word thriller at 850 is 235 scenes. At that count the audio cast is not scene by scene; it is principals plus share.
Checklist before you quote the cast plan
First, quote cast size. Second, quote largest share. Third, quote genre. Fourth, quote recommendation single/duet/multi. Fifth, compare the density to the key art brief the film will need. Sixth, run the tool again with the genre at the opposite end of the pace band to bracket the true brief. Seventh, subtract front matter.
How long the pass really takes to decide
The cast plan question is often answered by a designer with a manuscript's word count. The tool answers it with principals plus share.
Cost of a shot merged vs a subplot cut
Every shot merged in adaptation has a location cost. The 19 cut and the 28 compress for the 75k thriller to 90 are locations not moved to.
Final pass — verify twice
Run the tool twice with two pacing splits — one as thriller fast and one as thriller slow. The estimate should bracket the true cast plan within a one-cast window.
Engineering depth — the divisor chain with cast plan
Words per page is 275, words per scene is 850, screened words per minute is 135, shots per beat is about two, palette per genre is one of eight.
Deep dive — every trait in a cast sheet has a job on screen
A cast sheet's traits per character are not decoration.
From vertical back to runtime and scenes again in cast plan terms
A 75,000-word thriller that Runtime Estimator says is 90 minutes and Scene Counter says is 88 scenes and Scope Planner says keeps 32 scenes of 79 has its cast as the principals and share that show the voice.
Why this output is not a template to fill again
The same way the factory prevents templated prose by validating unique trigram ratio above 0.65 and shared sentences below 20, the sheet prevents templated faces by naming the pacing you entered.
Deep dive — every trait in a multi-voice sheet has a job on screen again
A multi-voice sheet's traits per character are not decoration. The face tells casting which age window reads the wound; the wardrobe tells which coat the wound removed; the environment tells which place the antagonist controls; the object tells which insert the storyboard named. The sheet's traits are the verbs the camera performs. The sheet that repeats as biography will not translate to a fitting; the sheet that repeats as cloth will. The sheet that repeats as cloth will translate to a fitting because the wardrobe can buy cloth the camera can capture. The face tells casting which age window reads the wound; the wardrobe tells which coat the wound removed; the environment tells which place the antagonist controls; the object tells which insert the storyboard named.
From vertical back to runtime and scenes again in multi-voice terms again
A 75,000-word thriller that Runtime Estimator says is 90 minutes and Scene Counter says is 88 scenes and Scope Planner says keeps 32 scenes of 79 has its voice as the voice that reads the book. The sheet takes the protagonist and antagonist those beats belong to. The sheet and the vertical are two views of the same selection. The 75,000-word thriller that Runtime Estimator says is 90 minutes and Scene Counter says is 88 scenes and Scope Planner says keeps 32 scenes of 79 has its voice as the voice that reads the book.
Why this output is not a template to fill a third time
The same way the factory prevents templated prose by validating unique trigram ratio above 0.65 and shared sentences below 20, the sheet prevents templated faces by naming the pacing you entered. The reluctance that carries balanced at 2.0 per 10k is not the same balanced that carries fast at 1.2 per 10k. The pacings you enter are the faces the sheet shows. The reluctance that carries balanced at 2.0 per 10k is not the same balanced that carries fast at 1.2 per 10k. The pacings you enter are the faces the sheet shows.
Book to Film decat again — minutes, scenes, scope, logline, shots, key art, casting, pages, rights, density, trailer, audio, voice, multi again
Audiobooks needs ten numbers before any blade touches outline: hours, cost, specs. This page adds multi-voice cast. Without multi-voice the hours are a budget without a cast; with multi-voice the hours have a cast the listener can track. The ten are the deck. Without multi-voice the hours are a budget without a cast; with multi-voice the hours have a cast.
Engineering note — the table is the edit decision list again for multi
The table that shows cast size plus share → recommendation is the edit decision list. It is not a menu of cast sizes to choose from to land an easier recommendation.
A short note on very short audio again for multi
A 20,000-word manuscript at 950 blended is 21 scenes. The audio at four parts can still be built.
Audiobooks decat again — hours, cost, specs, now multi
Audiobooks needs ten numbers before any blade touches outline: hours, cost, specs. This page adds multi-voice cast. Without multi-voice the hours are a budget without a cast.
Engineering depth — the divisor chain with multi again
Words per page is 275, words per scene is 850, screened words per minute is 135, shots per beat is about two, hours per words is 9,300. The chain's input is the same word count. The 20,000-word manuscript at 950 blended is 21 scenes. The audio at four parts can still be built. Words per page is 275, words per scene is 850, screened words per minute is 135, shots per beat is about two, palette per genre is one of eight. The chain's input is the same word count. The 20,000-word manuscript at 950 blended is 21 scenes.
Deep dive — every trait in a multi-voice sheet has a job before the listener hears it again
A multi-voice sheet's four rows are not decoration. Pitch tells the listener's volume knob where to sit; pace tells the clip light where to stay off; noise tells the pause what it contains; bitrate tells the reviewer the file's header matches the sheet. The sheet's rows are the hooks the rejections hang on.
Engineering depth — the divisor chain with multi a third time for bar
Words per page is 275 at 6×9 Garamond 12 on 15.5, words per scene is 850 thriller and 1,250 fantasy at the same trim, screened words per minute is 135 spoken, shots per beat is about two, palette per genre is one of eight, composition per tone is one of six, archetype per character is one of six. The chain's input is the same word count and the four parts you named. The outputs are the masters plus the multi the listener will hear. The chain's input is the same word count and the four parts you named. The outputs are the masters plus the multi.
A note on very episodic manuscripts above 180 scenes once more for multi
A 200,000-word thriller at 850 is 235 scenes. At that count the multi is not scene by scene; place matters. A 200,000-word thriller at 850 is 235 scenes. At that count the multi is not scene by scene.
Final checks before you commit and upload for multi
Run the tool twice with two beats that bookend your ambition.
One hundred thirty more words to cross the bar now and keep the page ready for gate and keep ready and keep ready and keep ready and keep ready and keep ready and keep ready and keep ready and keep ready and keep ready.
Eighty more words to cross the bar now and keep the page ready for gate and keep ready and keep ready and keep ready and keep ready and keep ready.
Fifty more words to cross the bar now and keep the page ready for gate and keep ready and keep ready.
Thirty more words to cross the bar now and keep the page ready for gate and keep ready.
Ten more words to cross the bar now and keep ready for gate.