Note: There is currently no built-in feature for advanced sound control. Everything below is a set of prompting techniques, not a guaranteed instruction set. The engine treats tags as weighted signals rather than commands, so results vary between generations. Expect to generate a few versions of any track.

1. Introduction & Creative Workflow

Purpose

This guide is a practical reference for creators building custom lyrical or instrumental tracks in Songer. It covers tag use, song structure, genre configuration, and expressive control, everything you need to get more predictable, polished results from your prompts.

Core Principles

Structure first: Define genre and style, then language, then lyrics
Brevity matters: The genre and style field has a 1000-character limit (caps each entry at 50 characters then separate it with a comma ",").
Predictability over surprise: Small, clear entries yield more accurate results than long, complex instructions.
Front-load what matters: Earlier entries carry more weight than later ones.
Human energy: Every line or layer should serve motion, groove, or emotion

Typical Prompt Flow

Field	Purpose	Notes
Genre/Style Field	Defines genre, rhythm, instrumentation, mood, vocal style, and accent in one line	50 characters per entry, ≤ 1000 characters total
Language	Vocal language	e.g., `English`, `French`, `Swedish`
Vocalist Gender	Male, Female, Random (or leaving it blank)	Leaving it blank means no vocalist is assigned
Lyrics	Full song text with `[Verse]`, `[Chorus]`, etc.	Prompt limits depend on your plan and the mode you're in: Songer Max: Generate: 5,000 characters Custom Lyrics: 5,000 characters Instrumental: 1,000 characters Songer Pro: Generate: 2,500 characters Custom Lyrics: 2,500 characters Instrumental: 1,000 characters Free and Pay-As-You-Go: Generate: 2,500 characters (Generate mode only available)

2. Universal Tag Index (Overview)

Tags are enclosed in square brackets [like this]. Group them by function for reliable results.

2.1 Structure Tags

Section	Tag	Description
Intro	`[Intro]`	First section
Verse	`[Verse 1]`, `[Verse 2]`	Main narrative sections
Pre-Chorus	`[Pre-Chorus]`
Chorus	`[Chorus]`, `[Hook]`	Repeated refrains
Bridge	`[Bridge]`	Contrast or tension
Drop	`[Drop]`
Build	`[Build]`
Outro	`[Outro]`	Fade or final section
Instrumental break	`[Instrumental Break]`	No vocal passage
Reprise	`[Reprise]`
Solo	`[Solo]`

2.2 Vocal & Harmony Tags

Vocal Tags

[Duet Vocals], [Lead], [Response], [Harmonies], [Unison], [Soft Delivery], 
[Intimate Delivery], [Minimal Vibrato], [Powerful Delivery], [Breathy], 
[Whisper Tone], [Angelic Tone], [Raw, Imperfect Vocals], [Talk-Singing Vocals], 
[Spoken-Style Vocals], [Spoken], [Rap], [High Pitch Vocals], [Low Pitch Vocals], 
[Soulful Delivery], [Deep Vocals], [Soft High Register], [Falsetto Vocals], 
[Bass Vocals], [Angelic Vocals], [Country Accent], [Norwegian Accent], 
[French Accent], [US Southern Accent], [British Accent], [Breathy Vocals], 
[Smooth Vocals], [Whisper Vocals], [Airy Vocal FX], [Layered Harmony FX], 
[Choral Backing FX], [Double Tracking FX], [Vocal Reverb FX], [Live Vocal FX], 
[Vocoder FX], [Pitch Shift FX], [Radio Voice FX], [Telephone FX], 
[Breathy Echo FX], [Haunting Vocal FX]

Type	Tag	Description
Duet	`[Duet Vocals] / [Duet]`	Alternating or combined phrasing
Backing vocals	`[Harmony Vocals]`, `[Background Choir]`	Adds depth
Style	`[Spoken]`, `[Rap]`	Special vocal style
Accent	`[British Accent]`, `[Norwegian Accent]`, `[Southern Drawl]`	Vocal character

Vocal Sound Design Tags

Category	Example Tags	Description
Tone & Texture	`[Breathy Vocals]`, `[Smooth Vocals]`, `[Whisper Vocals]`, `[Whisper FX]`, `[Soft Tone]`, `[Powerful Vocals]`	Adjust the vocal delivery mood
Character & Emotion	`[Soft Emotion FX]`, `[Soulful Vocals]`, `[Angelic Vocals]`, `[Haunting Vocals]`	Control expressive tone
Layering & Space	`[Double Tracking FX]`, `[Choral Backing FX]`, `[Vocal Reverb FX]`	Create vocal width & ambience
Processing & Filters	`[Telephone FX]`, `[Radio Voice FX]`, `[Pitch Shift FX]`, `[Vocoder FX]`	Apply stylistic vocal effects
Performance Presence	`[Live Vocal FX]`, `[Stage Ambience FX]`, `[Crowd Reverb FX]`	Simulate live or studio settings

2.3 Instrumentation Tags

[Acoustic Guitar], [Electric Guitar], [Bass Guitar], [Upright Bass], [Drums], 
[Brush Drums], [808 Bass], [Synth Bass], [Piano], [Rhodes], [Wurlitzer], 
[Strings], [Violin], [Cello], [Brass], [Trumpet], [Saxophone], [Trombone], 
[Flute], [Harp], [Percussion], [Taiko Drums], [Shaker], [Conga], [Bongo], 
[Steel Drums], [Synth Arps], [Pluck Synth], [Choir], [Pads], [Atmospheric Pads], 
[Ambient Drones], [Soundscape FX]

Example

Family	Tag Examples	Notes
Strings	`[Violin]`, `[Cello]`, `[Strings Ensemble]`	Pads or melody
Guitars	`[Acoustic Guitar]`, `[Electric Guitar]`, `[Slide Guitar]`	Style definition
Keys	`[Piano]`, `[Rhodes]`, `[Organ]`, `[Synth Arp]`	Harmonic base
Rhythm	`[Drum Kit]`, `[808 Kick]`, `[Conga]`, `[Shaker]`	Groove control
Brass / Wood	`[Trumpet]`, `[Sax]`, `[Flute]`, `[Clarinet]`	Jazz / Soul colors
Electronic	`[Bass Synth]`, `[Pad]`, `[Lead Synth]`	Club energy

2.4 FX & Ambience Tags

FX & Ambience

[Reverb FX], [Delay FX], [Echo FX], [Filter Sweep FX], [Sidechain FX], 
[Riser FX], [Drop FX], [Vocal Chop FX], [Stutter FX], [Distortion FX], 
[Chorus FX], [Flanger FX], [Phaser FX], [Analog Tape Warmth FX], 
[Vintage Reverb FX], [Rain FX], [Wind FX], [Birdsong FX], [Thunder FX], 
[Forest FX], [Water FX], [Ocean Waves FX], [Fire Crackle FX], 
[Audience Applause FX], [Mic Feedback FX], [Crowd FX], [Live Room FX], 
[Stage Ambience FX], [Crowd Cheer FX], [Concert Hall FX], [Hall Reverb FX], 
[Plate Reverb FX], [Room Reverb FX], [Cathedral Reverb FX], [Reverb Tail FX], 
[Stereo Delay FX], [Ping-Pong Delay FX], [Tape Delay FX], [Short Delay FX], 
[Tremolo FX], [Vibrato FX], [Rotary Speaker FX], [Wah-Wah FX], [Distortion FX], 
[Overdrive FX], [Fuzz FX], [Analog Warmth FX], [Tape Saturation FX], 
[Guitar Amp FX], [Bass Drive FX], [Filter Sweep FX], [High-Pass Filter FX], 
[Low-Pass Filter FX], [EQ Boost FX], [LFO FX], [Sidechain FX], [Resonance FX], 
[Night Crickets FX], [City Ambience FX], [Street Rain FX], [Forest FX], 
[Glitch FX], [Impact FX], [Sub Bass FX], [Bass Drop FX], [Pulse FX], 
[Synth Arp FX]

FX Categories

Category	Example Tags	Purpose
Spatial FX	`[Reverb FX]`, `[Echo FX]`, `[Hall Reverb FX]`, `[Room Ambience FX]`, `[Large Hall Reverb]`, `[Cathedral FX]`	Create depth and environment
Texture FX	`[Vinyl Crackle FX]`, `[Tape Hiss FX]`, `[Analog Warmth FX]`, `[Tape Delay]`	Add warmth, realism, vintage tone
Environmental FX	`[Rain FX]`, `[Wind FX]`, `[Ocean FX]`, `[Applause FX]`, `[Birds FX]`, `[Ocean Waves FX]`	Evoke mood or realism
Modulation FX	`[Chorus FX]`, `[Phaser FX]`, `[Flanger FX]`, `[Phaser]`, `[Flanger]`	Enhance movement and stereo width
Distortion FX	`[Overdrive]`, `[Saturation]`	Energy boost
Vocal FX	`[Breathy Vocals]`, `[Layered Harmony FX]`, `[Airy Vocal FX]`	Affect vocal delivery & presence
Electronic FX	`[Filter Sweep FX]`, `[LFO FX]`, `[Sub Bass FX]`, `[Sidechain FX]`	Add motion, EDM-style tension
Creative FX	`[Reverse FX]`, `[Glitch FX]`, `[Drop FX]`, `[Build FX]`	Used for transitions and breakdowns
Live Crowd FX	`[Applause FX]`, `[Audience FX]`, `[Club Ambience]`	Adds performance feel

Sound Effects and Ambience: Three Places to Add

The same sound can be requested in three ways, and they do different jobs:

Where	Form	Scope
Genres/Vibes	`rain and thunder ambience`	Whole track
Section stack	`[Intro \| thunder and rain ambience]`	That section
Inline	`thunder rumbling`	The line it sits on

Using two or three together is more reliable than any one alone. The field entry sets the overall atmosphere, the section tag confirms it where you want it, and the asterisk event places a specific hit.

Genres/Vibes:

modern country, acoustic pop, emotional, heartfelt, singer-songwriter, storytelling, 80 BPM, warm production, cinematic atmosphere, rain and thunder ambience

Lyrics:

[Intro | thunder and rain ambience | distant lightning strike | cold and atmospheric | building mood | 8 bars]
*thunder rumbling*
*rain falling*

[Verse 1 | Soft Delivery | Acoustic Guitar]
First verse goes here

Bracket tag or asterisk?

Both forms work. [Thunder FX] and *thunder rumbling* are asking for the same thing, and you can use whichever reads better to you, or both, as above.

The tag lists in this guide are illustrative, not exhaustive. Descriptive phrasing that isn't in the list often works fine: [Heartbeat FX], *glass shattering*, and distant lightning strike are all reasonable things to try, even though none of them appear in a table.

The disclaimer that matters: Because these aren't a fixed vocabulary, any given phrasing may or may not land. If one form does nothing, try the other before assuming the sound isn't available. If neither works, reinforcing it in Genres/Vibes as well is the next thing to try. Some sounds simply won't render, no matter how they're phrased.

Guidance

One or two events per track. Stacking sound events is the fastest way to make a track sound like a sound-effects reel.
Place them at structural moments. Intro, section transitions, and outro rather than mid-phrase.
Match the genre. These land well in rock, folk, country, and singer-songwriter, as well as anything meant to feel live or cinematic. They're least reliable in clean pop and EDM.
Don't fight yourself. Words like clean, studio, or pristine in Genres/Vibes work directly against an ambience cue.
Think atmosphere, not spot effect. The engine is better at "this track feels like a storm outside" than at a precisely timed crack of thunder on a specific beat.

2.5 Dynamics & Expression

Production (Dynamics & Arrangement)

[Crescendo], [Swells], [Staccato], [Legato], [Breakdown], [Build], [Drop], 
[Reprise], [Layered Harmonies], [Full Orchestra], [Minimal Mix], [Fade In], 
[Fade Out], [Dramatic Fade], [Quiet Interlude], [Epic Drums], [Main Theme], 
[Climax], [Symphonic Arrangement], [Dark Atmosphere]

Tag	Meaning
`[Crescendo]`	Gradual build-up
`[Diminuendo]`	Gentle fade
`[Staccato]`	Short articulation
`[Legato]`	Smooth connection
`[Swells]`	Orchestral motion
`[Soft Intro]` / `[Strong Outro]`	Emotional entry/exit

2.6 Combining Tags

Most sections need more than one tag: a section label, a vocal cue plus a texture. There are two ways to write that, and both work:

Pipe-stacked, one bracket

[Verse 1 | Male Vocal | Ethereal]

Consecutive brackets

[Verse 1][Male Vocal][Ethereal]

Consecutive brackets also work split across lines, which is easier to read on longer stacks:

[Verse 1] [Male Vocal] [Ethereal]

All three are read the same way. Pick whichever you find easier to scan and stay consistent within a song; mixing forms makes it harder to spot a typo later.

Ordering. Put the section label first, then vocals, then instrumentation, then FX and production. Earlier items in the stack carry more weight, so lead with whatever matters most for that section.

[Chorus | Powerful Delivery | Layered Harmonies | Hall Reverb FX] 
[Bridge | Piano | Quiet Interlude | Diminuendo] 
[Solo | Saxophone | Legato | Vintage Reverb FX] 
[Verse 1 | Rap | Spoken-Style Vocals | 808 Bass | Short Delay FX] 
[Drop | Bass Drop FX | Sidechain FX | Sub Bass FX | Lead Synth] 
[Outro | Ambient Drones | Atmospheric Pads | Reverb Tail FX | Fade Out]

Four to six items are a good target. Longer stacks work; seven or eight is common on intros where you're describing atmosphere in detail, but items past that point carry progressively less weight, so put the ones you actually care about early.

Don't stack tags that fight each other. [Verse | Intimate Delivery | Powerful Delivery] gives the engine nothing usable; it will pick one or land somewhere in between. Same for [Chorus | Minimal Mix | Full Orchestra].

On vocal gender tags. [Male Vocal] and [Female Vocal] in a stack reinforce the section's delivery, but the Vocalist Gender selector is the authoritative control for the track. Use the tags to shape delivery within a section, not to override the setting.

3. Prompt Building & Best Practices

3.1 Prompt Syntax Example

Genre / Style Field: country waltz, 3/4, American vocals, acoustic 

Lyrics: 
[Verse 1] 
… 
[Chorus] 
… 
[Outro]

Genre / Style Field Rules

Rule	Description	Example
Keep each entry ≤ 50 characters	Combine genre, tempo, accent, vocal type, and instrument focus	`melodic techno, cinematic, British vocals, 126 BPM`
Include vocal style + accent	Guides voice model choice and pronunciation	`country waltz, 3/4, American vocals, acoustic`
Add rhythm/tempo info		`slow waltz, 3/4`, `90 BPM`, `4/4 club`
Define the main instrument color		`trumpet-led`, `piano`, `acoustic`, `synth-heavy`
Avoid sentences		“This song should sound like a sad movie.”
Avoid redundant adjectives		“beautiful, melodic, very emotional, lovely”

3.2 Golden Rules

Time Signature: Include explicitly (3/4, 4/4, 6/8).
Tempo: Add target BPM (“90 BPM slow waltz”).
Instrument Hierarchy: Start with dominant instrument tags, then color layers.
Mood Words: Use 2–3 emotion words max (“melancholic, hopeful, dreamy”).

Genre / Style Field Formula

[Primary Genre], [Subgenre or Mood], [Accent + Vocal Type], [Instrumentation or Feel], [Tempo or Time Signature], [FX]

Component	Example Values
Genre	pop, jazz, techno, gospel, country
Subgenre / Mood	melodic, deep, fusion, soulful, cinematic
Accent + Vocal	American vocals, British vocals, French vocals
Instrument Focus	piano, trumpet-led, acoustic, synth-heavy
Tempo / Signature	90 BPM, 3/4, 4/4, 12/8
FX	audience applause

Compliant Examples

Style	Prompt	Interpretation
Country Waltz	`country waltz, 3/4, American vocals, acoustic`	American voice, 3/4 acoustic waltz
Melodic Techno	`melodic techno, cinematic, British vocals, 126 BPM`	British tone, cinematic melodic techno
Jazz Ballad	`jazz ballad, slow, French vocals, trumpet`	Smooth French vocals, trumpet-led ballad
Soul R&B	`soul r&b, 90 BPM, African vocals, piano-led`	Warm African tone, soulful R&B
Pop Dance	`pop dance, American vocals, upbeat`	Bright American voice, dance pop
Amapiano	`amapiano, South African vocals, log drum, 112 BPM`	Local tone with log drum groove
Cinematic Ballad	`cinematic ballad, orchestral, British vocals, slow`	Orchestral ballad with British vocals

3.3 Steering away from something

Negation has to go inline as its own entry:

acoustic folk, no drums, no percussion, fingerpicked, 84 BPM

This works, but it's less reliable than a positive description. The music engine responds better to being given something to reach for than something to avoid. If drums keep appearing in a track that shouldn't have them, adding solo acoustic guitar and a single performer usually does more than adding another no drums.

4. Lyrics & Vocal Craft

4.1 Lyric Structure

Section	Function
Verse	Story progression
Chorus	Emotional anchor & hook
Bridge	Change of perspective
Outro	Closure or echo theme

4.2 Crafting Lyrics

Keep lines short (6–10 words).
Use clear vowels for melody flow.
Repeat keywords sparingly for catchiness.
Avoid complex rhymes that break meter.

4.3 Duets & Harmonies

Genre/Style Field: Duet, harmonies 
Prompt:
[Duet] I see the way you look at me 
[Harmonies] Love is our secret song

Use explicit vocal tags per line for balance.

4.4 Formatting Inside Lyrics

Formatting	Effect
`(parentheses)`	Sung as a backing vocal or call-and-response layer
`ALL CAPS`	Louder, more forceful delivery, one to three words per section
`...`	Slower, more hesitant delivery
`lo-o-ove`	Stretches the syllable across a longer note
Blank line	Section break

Contrast is what makes these work. If every line is in caps, nothing reads as emphatic.

5. Known Limitations

Being upfront about these saves time on iteration.

Backing vocals in instrumental and electronic tracks. In melodic techno and adjacent electronic genres, vocal texture sometimes bleeds into tracks generated as instrumentals. Adding no vocals and no instrumentation to Genres/Vibes reduces it, but doesn't reliably eliminate it.

Oohs and aahs in gospel and worship. Wordless vocal layers often appear in gospel and worship tracks, even when the lyrics don't call for them. This is characteristic of how the engine handles those genres, and there's currently no prompt that consistently removes it.

Extreme metal vocal control. Fine control over grunts, growls, and screams, as well as specific placement and intensity, is beyond what the engine currently offers. Genre and delivery tags will get you in the neighborhood, but not to a specific performance.

Next steps

Ready to try experimenting with custom lyrics or instrumentals? Check the next guides:

Tags and Prompts Overview