Audio Prompts

Music briefs with looping specs and leitmotif seeds enable consistent world-building through audio. The loop point specification prevents the jarring mid-note restart that ruins overworld music.

Adaptive combat music that layers based on game state creates subconscious stress that players feel but don't consciously notice — exactly what great combat audio does.

Story scene music needs explicit dynamic arcs and sync points to match cutscene timings. The "how to edit for sync" section bridges the gap between generation and integration.

Title screen music that introduces the leitmotif trains players to recognize it throughout the game — making those later moments of recognition deeply satisfying.

Ambient tracks with three danger-state variations enable adaptive audio systems that react to gameplay without expensive custom implementations.

Menu music systems that share harmonic language create unconscious world coherence. The "doesn't compete with reading" requirement for inventory screens prevents the music that causes players to mute everything.

Boss music without phase-aware design misses the chance to make health thresholds feel dramatic. The "music breaks in some way" requirement in phase 3 is what makes final phases memorable.

Exploration music that doesn't communicate the region's identity misses its most important job. The stem-layer brief creates the dynamic build that rewards thorough exploration.

Town themes play the most hours of any track in the game. Designing for ear fatigue tolerance and subtle motif placement is what separates good town music from great.

Cutscene music without a timestamp-based brief results in music that's beautiful but doesn't serve the story beats. Sync points are non-negotiable for emotional landing.

Credits music that's just a "happy version of the main theme" wastes the emotional resonance of the ending. The transformed leitmotif requirement is what makes players sit through the credits.

Main menu themes heard hundreds of times require long-relationship design. The idle variation prevents the tune from becoming actively grating during session setup.

Puzzle music that creates unconscious time pressure frustrates players. Explicitly designing what is removed — not just what is present — is what makes puzzle music support cognition rather than compete with it.

Stinger libraries designed without emotional specificity end up as generic "good" and "bad" jingles. Eight distinct emotional profiles give composers a palette that fits actual story moments.

Leitmotifs are the musical stitching of a narrative score. The diegetic appearance is the touch that makes players pause and recognize the theme in a new context — a moment of genuine joy.

Generative music systems sound algorithmic when variation decisions are made without harmonic constraints and human-authored source material. The "composer in the loop" rule prevents the system from replacing the human entirely.

Emotional climax music that swells conventionally becomes background noise. The "one note not to play" and the "break convention" requirements force the unexpected choice that creates genuine emotional impact.

Combat music that switches tracks rather than layering creates jarring cuts the player hears as glitches. A stem-based layering system that follows combat state is the professional implementation — and this brief documents it before the composer starts.

Main menu music without a deliberate emotional target defaults to generic epic orchestral. The foreshadowing motif requirement connects the first sound the player hears to the emotional climax — a technique that players feel even when they don't notice it.

SFX specs with "hit feel punch layers" — the non-realistic elements — acknowledge that game audio is always designed fiction, not reality. These layers are what separate satisfying from flat combat.

UI SFX specs with emotional intent per sound prevent the "same click sound on everything" design that makes a UI feel flat. The channel budget prevents UI sounds from drowning gameplay audio.

Three-layer ambience systems that separate constant, periodic, and reactive sounds are significantly easier to tune than flat ambience beds — and produce richer results with the same asset count.

Weapon sounds without a layering brief produce thin audio that doesn't match visual impact. The frequency emphasis spec prevents the "my gun sounds like a toy" problem.

UI sounds designed without a consistent identity feel like they were pulled from different games. The "3 adjectives" anchor ensures every sound is judged against the same standard.

Single-track ambience loops become invisible to the ear within minutes. A layered ambience system with randomized periodic sounds maintains perceived freshness indefinitely.

Footstep systems without sufficient variation create the "heartbeat" effect — players hear the same sound pattern and the loop becomes intrusive. Four variations per foot per surface is the minimum.

Spell audio without a "sonic family" rule sounds like random sound effects. The real-world source reference grounds the synthesis direction and creates recognizable identity across an entire magic school.

Death audio sets the emotional tone for how players experience failure. The silent death vs musical sting decision shapes whether death feels dramatic or functional.

Crafting audio without tiered success sounds misses a major player feedback opportunity. Three distinct success sounds that stay in family teach players what quality they achieved without a tooltip.

Level-up sounds heard 100 times become intrusive if not designed for longevity. The "variation strategy" and "level 100 vs level 2" questions force the design that prevents the most over-heard jingle in the game.

Currency pickup sounds heard hundreds of times become the most annoying audio in the game if not designed with anti-fatigue in mind. The explicit anti-fatigue design choice requirement forces this consideration.

Footstep systems specified without minimum variation counts produce the "wooden plank" problem: 3-4 sounds cycling detectably. 6-8 variations per surface with pitch and volume randomization is the threshold where repetition detection stops.

Voice direction sheets with "who they are NOT" sections prevent voice actors from defaulting to the most obvious read of a character type. The essence-capturing line gives actors something to anchor every other line to.

Combat barks that are scripted in volume with variety flags prevent the repeated-line fatigue that makes players mute characters. The "no two lines start the same word" rule forces real vocal variety.

AI voice production workflows for indie games that specify when to keep vs re-generate prevent teams from spending hours regenerating lines that a human reviewer would pass. The stability recommendation is the least-understood ElevenLabs setting.

Remote voice recording without clear direction produces expensive re-records. Sample lines with labeled interpretations are the single most effective alignment tool.

Combat barks that sound generic undermine character identity. The "reveals backstory" requirement turns a bark line into world-building in 5 words or less.

Villain voice direction that defaults to "deep and threatening" produces generic antagonists. Specifying the villain's controlled emotional register produces memorably cold menace.

Player character casting is uniquely difficult — too much personality alienates players, too little produces a blank slate. The identity target section forces the team to define this exact balance before auditions.

Narrators who sound neutral rather than specific become invisible — players stop registering the story. A specific emotional register and tested sample during casting creates a voice that defines the game.

Ensemble casts without a vocal diversity matrix produce characters indistinguishable in cutscenes. The comparative table reveals duplicate patterns before casting rather than after recording.

Ambient NPC voices with generic lines produce the atmosphere-destroying "WHAT ARE YOU BUYING" effect. Banning stock phrases and setting a 3-8 word limit forces world-specific writing that makes the environment feel real.

Audio naming conventions that align with designer, programmer, and audio middleware terminology prevent the "what is this event called again" friction that slows audio iteration.

Adaptive music systems without priority logic create undefined behavior when multiple states trigger together. Defining the priority matrix before implementation prevents audio bugs that are invisible in code but obvious to players.

Most indie developers mix by feel with no reference — this guide gives them a systematic process. The "reference track" method produces dramatically better results than mixing in isolation.

Audio middleware architecture designed by sound designers without programmer input creates integration disasters. This spec bridges both disciplines.

Music transitions planned in isolation produce jarring cuts. The full matrix reveals contradictions (e.g., combat→death should be instant, but combat→victory needs space to breathe).

Spatial audio designed without occlusion feels two-dimensional even in a 3D world. The wall-passthrough behavior is what separates games that feel sonically real from those that don't.

Adaptive music that reacts too quickly to gameplay events sounds like a skipping CD. The intentional lag design — built in from the start — prevents the most common adaptive music failure.

Dialogue implementation without a naming convention produces an unmanageable audio library at scale. The naming convention is the most impactful technical decision — get it wrong and it costs hours per patch.

Audio reverb zones without transition crossfades produce jarring acoustic snaps when crossing boundaries. The transition duration spec is a small detail that completely changes the perceived audio quality.

Audio optimization decisions made per platform prevent the most common port failure: a game that sounds great on PC and sounds broken on Switch because streaming wasn't considered at design time.

Audio without occlusion makes every sound feel like it's in the same room as the player regardless of walls. Specifying the HPF cutoff and send routing for occluded sounds before implementation prevents the flat, papery audio mix that requires expensive retrospective fixes.