BUILD THE LOADZ AUDIOBOOK PREVIEW EXPERIENCE
Create a premium audiobook feature for the first two chapters of LOADZ.
This is not a temporary text-to-speech demo.
Generate, approve, save, and securely host permanent audio files so every listener hears the exact same approved performance every time.
The audio must:
• resume from the same timestamp
• pause and restart without regenerating
• support seeking
• preserve the narrator’s approved performance
• work across desktop and mobile
• connect naturally to THE FOUNDATION AI
• allow visitors to hear Trucker J. guide them into the world of LOADZ
The goal is to make the opening chapters feel professionally narrated, culturally authentic, emotionally alive, and worthy of a commercial audiobook release.
==================================================
CREATIVE STANDARD
Create the emotional quality of a premium Audible-style audiobook without copying or impersonating any identifiable narrator, actor, celebrity, or real person.
Do not clone a real person’s voice.
Do not imitate a specific living narrator.
Instead, create an original fictional narrator voice with these qualities:
• Black American male
• approximately 38 to 48 years old
• raised in or deeply familiar with Cleveland, Ohio
• warm baritone
• grounded
• mature
• observant
• emotionally intelligent
• naturally street-aware
• confident without performing toughness
• capable of humor, tension, tenderness, and danger
• believable as someone who has driven long roads, raised a family, survived pressure, and learned when silence matters
The voice should carry the emotional energy associated with Trucker J.:
• provider
• protector
• husband
• father
• working man
• patient observer
• quiet authority
• somebody who thinks before he speaks
The narrator is inspired by Trucker J.’s emotional qualities.
The narrator must not claim to literally be Trucker J. unless the website clearly labels the performance as an in-character fictional reading.
==================================================
PUBLIC FEATURE NAME
Use:
HEAR THE STORY
Supporting line:
The road sounds different when somebody who knows the weight is telling it.
Alternative section title:
LISTEN TO LOADZ
Do not call it:
AI Voice Demo
Text-to-Speech
Generated Audio
Robot Narration
Audio Test
A small disclosure may appear beneath the player:
This preview uses an original AI-assisted performance created specifically for the LOADZ experience.
==================================================
CONTENT SCOPE
Create audiobook previews for:
LOADZ
Chapter One
Chapter Two
Use the final approved manuscript text exactly as written.
Do not:
• rewrite sentences
• clean up dialogue during narration
• change dialect
• replace slang
• add dialogue
• remove profanity
• summarize scenes
• alter character names
• censor language unless the author approves a separate clean version
• change story facts
• improvise beyond the manuscript
The audio should preserve the author’s wording while using performance, pacing, silence, and character distinction to bring it alive.
Before generating audio, confirm that the Chapter One and Chapter Two manuscripts are the approved final versions.
If multiple manuscript versions exist, stop and ask which version is canonical.
==================================================
NARRATOR VOICE PROFILE
Create one original narrator voice and preserve it as the permanent LOADZ narration identity.
VOICE CHARACTERISTICS
Pitch:
Low to medium-low baritone.
Not unnaturally deep.
Not a movie-trailer growl.
Tone:
Warm, textured, lived-in, controlled.
Cadence:
Natural Great Lakes and Black Midwestern rhythm.
The cadence may carry traces of Cleveland familiarity, but it must never become a caricature.
The narrator should sound like a real Black man from Cleveland telling a story to people who understand the city.
He should not sound like:
• a generic Southern narrator
• a New York stereotype
• an exaggerated “street” performance
• a preacher throughout the entire reading
• a radio announcer
• a commercial voice-over
• a rapper reading lyrics
• a motivational speaker
• an actor forcing slang
• an older man trying to sound young
• a young man trying to sound hardened
PRONUNCIATION
Respect:
• Cleveland neighborhood names
• local streets
• Black names
• Latino names
• trucking terminology
• vehicle names
• street slang
• regional phrasing
• character nicknames
• contractions
• intentionally nonstandard grammar
Create a pronunciation dictionary before production.
Every uncertain proper noun must be reviewed by the author.
Do not guess repeatedly across different chapters.
==================================================
NARRATION PERFORMANCE
The narrator should read prose and dialogue differently.
PROSE
• calm but cinematic
• clear without sounding overly formal
• measured enough to absorb detail
• slightly slower during reflection
• tighter during action
• never monotone
• never overacted
DIALOGUE
Give every character a distinct emotional posture without turning the reading into comedy or impersonation.
Differentiate characters through:
• tempo
• confidence
• breath
• sentence attack
• pause length
• emotional pressure
• vocal placement
Do not create extreme cartoon voices.
Do not use exaggerated accents.
Do not make women sound like parodies.
Use subtle performance changes.
==================================================
TRUCKER J. PERFORMANCE
Trucker J.’s dialogue should feel:
• calm
• masculine
• thoughtful
• responsible
• tired when the road has been long
• softer around Toya and the boys
• cautious around unfamiliar people
• firm when protecting his family
• internally conflicted when pressure begins building
He does not waste words.
He should sometimes pause before answering, as if calculating the weight of what he is about to say.
He should never sound ignorant.
He is a hardworking professional with street awareness, emotional intelligence, and real responsibilities.
==================================================
TOYA PERFORMANCE
Toya should sound:
• confident
• warm
• quick when joking
• direct when serious
• affectionate without becoming weak
• naturally beautiful in presence, not performed through a breathy voice
• grounded as a wife and mother
• capable of taking control of a room
Do not make Toya sound submissive, shrill, or overly dramatic.
==================================================
RICO PERFORMANCE
Rico should sound:
• charismatic
• forceful
• neighborhood-rooted
• confident
• capable of humor
• dangerous when disrespected
• respected without constantly yelling
His volume should not be his personality.
His certainty should carry the authority.
==================================================
GUNNER PERFORMANCE
Gunner should sound:
• lower-energy than Rico
• precise
• restrained
• observant
• emotionally guarded
• calculated
• capable of making a quiet sentence feel heavier than shouting
==================================================
CHI-CHI PERFORMANCE
Chi-Chi should sound:
• friendly
• measured
• intelligent
• bilingual or culturally fluent where the manuscript supports it
• professional
• more dangerous because of his calmness, not because of a forced accent
Do not exaggerate Mexican or Latino speech patterns.
==================================================
PACING SYSTEM
Use intentional pacing rather than one fixed reading speed.
GENERAL TARGET
Average narration:
Approximately 145 to 165 spoken words per minute.
Reflective passages:
Approximately 125 to 145 words per minute.
Normal dialogue:
Approximately 150 to 175 words per minute.
High-tension action:
Approximately 165 to 185 words per minute, while preserving clarity.
Never rush important names, emotional revelations, or scene-setting details.
==================================================
PAUSES
Use silence as part of the performance.
Suggested pause logic:
Comma:
Brief natural breath, not always a full stop.
Period:
Approximately 250 to 500 milliseconds depending on emotional weight.
Paragraph break:
Approximately 500 to 900 milliseconds.
Scene shift:
Approximately 1 to 1.8 seconds.
Major revelation:
Approximately 800 milliseconds to 1.5 seconds.
Chapter title:
Allow 1.5 to 2 seconds after the title before narration begins.
Do not apply these mechanically.
Adjust them to the emotional meaning of the sentence.
==================================================
EMOTIONAL DIRECTION
The narrator must understand the emotional function of each scene.
Before rendering, annotate the manuscript privately with:
• scene objective
• dominant emotion
• secondary emotion
• tension level
• character power balance
• relationship temperature
• pace
• required silence
• pronunciation notes
• transition notes
Suggested tension scale:
1:
Safe and domestic
2:
Curious
3:
Uneasy
4:
Suspicious
5:
Pressured
6:
Threatening
7:
Dangerous
8:
Immediate crisis
9:
Violent or emotionally explosive
10:
Major irreversible moment
Do not read every chapter at the same emotional intensity.
Contrast is essential.
The family scenes must feel warmer so the danger carries more weight.
==================================================
ENUNCIATION
The narrator should be clear without sounding polished beyond the world of the book.
Preserve natural speech features:
• contractions
• dropped endings where written
• rhythmic emphasis
• conversational compression
• authentic pauses
• local phrasing
• culturally natural stress
Do not pronounce every word as though reading a legal document.
Do not correct dialect through performance.
Do not over-pronounce slang.
The listener should understand every sentence while still feeling like the story belongs to its people.
==================================================
CLEVELAND SOUND AND CULTURAL TEXTURE
The reading should feel connected to Cleveland through:
• verbal rhythm
• emotional restraint
• blue-collar realism
• neighborhood familiarity
• weather awareness
• road and trucking references
• East Side and West Side distinctions where supported
• family-centered humor
• practical communication
Do not repeatedly say “Cleveland” to manufacture authenticity.
Authenticity should come from how the people speak, move, and respond.
==================================================
AUDIO PRODUCTION WORKFLOW
Do not generate the full chapters in one uncontrolled pass.
Use this production process:
- Lock the final manuscript.
- Divide each chapter into scenes.
- Identify narrator and character dialogue.
- Build a pronunciation dictionary.
- Assign emotional direction to each scene.
- Generate short test samples.
- Review voice consistency.
- Review pronunciation.
- Review pacing.
- Review character distinction.
- Correct individual lines where necessary.
- Render each approved scene.
- assemble the chapter.
- normalize audio.
- master the chapter.
- export permanent audio files.
- create transcript timestamps.
- test the web player.
- receive author approval.
- publish.
Do not publish the first generated take automatically.
==================================================
PERMANENT AUDIO REQUIREMENT
The final approved audiobook preview must be pre-rendered and saved as permanent audio files.
Do not regenerate the audio every time a visitor presses Play.
This guarantees:
• identical performance
• identical pacing
• identical pronunciation
• consistent timestamps
• reliable pausing
• reliable resuming
• predictable cost
• consistent sound quality
• stable chapter markers
Create separate master files:
loadz-chapter-01-master.wav
loadz-chapter-02-master.wav
Create optimized streaming files:
loadz-chapter-01.mp3
loadz-chapter-02.mp3
Optional modern format:
loadz-chapter-01.m4a
loadz-chapter-02.m4a
Use WAV as the archival master.
Use MP3 or AAC/M4A for browser playback.
==================================================
AUDIO QUALITY
Master specifications:
• clean voice recording
• no clipping
• no digital distortion
• no obvious AI glitches
• consistent loudness
• controlled sibilance
• minimal room coloration
• no background hiss
• no abrupt cuts
• natural breaths preserved selectively
• breaths reduced only when distracting
• seamless edits
Suggested final loudness target:
Approximately -18 to -16 LUFS integrated for spoken-word playback.
Suggested true peak ceiling:
Approximately -1 dBTP.
Keep chapter loudness consistent.
Do not crush the voice with heavy compression.
==================================================
BACKGROUND SOUND
Narration should remain the priority.
Optional atmospheric sound may be used sparingly.
Possible intro texture:
• distant truck idle
• wet-road ambience
• soft Cleveland wind
• faint warehouse room tone
• page turn
Possible scene transitions:
• brief road noise
• distant city ambience
• soft household atmosphere
• restrained restaurant ambience
Rules:
• never place constant music beneath the entire chapter
• never overpower dialogue
• never use copyrighted commercial music
• never add gunshots or violent effects merely for excitement
• do not turn the audiobook into a radio drama unless separately approved
• allow listeners to disable enhanced ambience if two mixes are produced
Preferred initial release:
Narration-forward mix with only a short opening and closing sound signature.
==================================================
AUDIO PLAYER PLACEMENT
Add the player prominently on:
/books/loadz
Place it near:
Begin Reading Free
Use a paired choice:
READ THE FIRST TWO CHAPTERS
LISTEN TO THE FIRST TWO CHAPTERS
Also place a compact player inside the free-reading experience.
Suggested heading:
HEAR THE ROAD OPEN.
Supporting copy:
Listen to the first two chapters of LOADZ, performed in an original voice created for the world of the book.
==================================================
PLAYER DESIGN
Create a custom, accessible audiobook player.
REQUIRED CONTROLS
• Play
• Pause
• Restart
• Skip Back 15 Seconds
• Skip Forward 15 Seconds
• Previous Chapter
• Next Chapter
• Seek Bar
• Current Time
• Total Duration
• Playback Speed
• Volume
• Mute
• Transcript
• Chapter List
• Download only if approved
• Close or minimize where applicable
PLAYBACK SPEEDS
0.75×
0.9×
1×
1.1×
1.25×
1.5×
1.75×
2×
Default:
1×
Do not alter pitch when changing speed.
==================================================
PLAYER VISUAL DESIGN
The player should match the LOADZ website.
Use:
• charcoal
• deep black
• warm ivory
• restrained red accents
• brushed-metal or paper-inspired texture
• readable waveform or progress line
• clear typography
• large controls
• subtle motion
Do not use:
• generic browser-player styling
• neon waveform
• spinning records
• cassette imagery
• cartoon headphones
• excessive visualizers
• nonstop animation
Suggested layout:
LOADZ
CHAPTER ONE
[Chapter Title]
[15 sec back] [Play/Pause] [15 sec forward]
00:00 ━━━━━━━━━━━━━ 18:42
1× Chapters Transcript Volume
Below:
Narrated in an original performance created for the LOADZ experience.
==================================================
PROGRESS BAR
The progress bar must:
• show buffered audio
• show current playback position
• support click-to-seek
• support drag-to-seek
• support keyboard seeking
• show elapsed time
• show total time
• remain readable on mobile
• update smoothly
• not reset during ordinary page interactions
Use a larger invisible touch target around the visible bar.
==================================================
PLAYBACK PERSISTENCE
When a listener pauses:
Store the exact timestamp.
When they return:
Offer:
CONTINUE LISTENING
Chapter One
12:43 remaining
Resume from the saved position.
For logged-out users:
Store progress locally on that device.
For signed-in members:
Store progress securely in their account so it can continue across devices.
Store:
• book
• chapter
• timestamp
• total duration
• playback speed
• last listened date
• completion percentage
Do not reset progress unless the listener chooses Restart.
==================================================
MINI PLAYER
When audio continues while the visitor navigates within the website, display a compact persistent player.
Desktop:
Bottom edge or lower corner.
Mobile:
Bottom bar above safe areas.
Include:
• chapter name
• play or pause
• 15-second rewind
• progress
• expand player
• close
Do not restart the audio during internal page navigation.
Use a global audio state or persistent layout.
==================================================
CHAPTER SELECTION
Create a chapter list:
CHAPTER ONE
[Approved Chapter Title]
Duration
Play
CHAPTER TWO
[Approved Chapter Title]
Duration
Play
Show:
• current chapter
• completed chapter
• in-progress chapter
• locked future chapters where appropriate
Do not imply the full audiobook is available unless it exists.
Future chapters may show:
Full audiobook coming later
only if that release is genuinely planned.
==================================================
TRANSCRIPT SYNCHRONIZATION
Create an optional transcript view.
Requirements:
• exact manuscript text
• current paragraph highlighted
• click paragraph to seek
• auto-scroll optional
• user can disable auto-scroll
• accessible screen-reader structure
• no forced transcript display
• chapter and scene markers
• preserve punctuation
Do not expose private production annotations.
==================================================
ACCESSIBILITY
The player must support:
• keyboard controls
• visible focus
• screen readers
• clear labels
• large tap targets
• transcript
• high contrast
• reduced motion
• no sound autoplay
• no audio required to access the written chapters
• captions or transcript for atmospheric introduction where relevant
Suggested keyboard shortcuts:
Space:
Play or Pause
Left Arrow:
Back 5 seconds
Right Arrow:
Forward 5 seconds
Shift + Left Arrow:
Back 15 seconds
Shift + Right Arrow:
Forward 15 seconds
M:
Mute
T:
Open Transcript
Do not activate shortcuts while the user is typing in a form.
==================================================
AUTOPLAY
Do not autoplay chapter narration.
The visitor must press Play.
A muted visual invitation may appear.
Do not begin speaking when the page loads.
Do not surprise visitors with audio.
==================================================
THE FOUNDATION AI CONNECTION
Connect the audiobook experience to THE FOUNDATION.
On the LOADZ page, suggested AI actions may include:
• Read Chapter One to Me
• Continue My Audio
• What Did I Just Hear?
• Who Was Introduced?
• Explain This Without Spoilers
• Talk About Trucker J.
• Switch to Chapter Two
When the visitor asks:
“Read Chapter One to me.”
THE FOUNDATION should open the permanent audio player and begin playback only after clear user interaction.
Do not synthesize a new performance on demand when the approved chapter recording exists.
Use the saved master audio.
==================================================
TALKING TO TRUCKER J.
Create a clearly labeled optional fictional character experience:
TALK WITH TRUCKER J.
Supporting disclaimer:
A fictional AI experience grounded in the character and world of LOADZ.
This experience must remain separate from the narrator identity.
The audiobook narrator tells the story.
The Trucker J. conversation responds as an approved fictional interpretation of the character.
Do not blur them together without labeling.
Trucker J. conversational qualities:
• calm
• protective
• reflective
• family-focused
• practical
• measured
• road-wise
• unwilling to spoil the story
• capable of light humor
• aware that choices carry consequences
He may discuss:
• trucking life
• family
• being away from home
• responsibility
• Cleveland
• discipline
• early-story experiences
• themes in LOADZ
He must not reveal:
• hidden plot information
• future betrayals
• deaths
• major Book 2 outcomes
• information he would not know at that point in the story
Suggested opening:
TRUCKER J.:
Long road ahead.
You can listen to the first two chapters, ask me about life behind the wheel, or learn what it means to provide when everybody is counting on you.
Just know one thing.
I’m not giving away the whole route before you take the ride.
==================================================
TRUCKER J. VOICE EXPERIENCE
If voice conversation is enabled later, use an original voice consistent with the approved audiobook identity.
Do not clone a real person.
Do not imitate an identifiable narrator.
Clearly indicate that this is a fictional AI character experience.
Require user interaction before microphone access.
Provide:
• Start Voice Conversation
• Mute Microphone
• End Conversation
• Show Transcript
• Switch to Text
Do not record voice conversations without clear consent.
==================================================
AUDIO RIGHTS AND APPROVAL
Before publishing, confirm:
• James owns or controls the preview rights
• the final manuscript is approved
• the narrator voice is original or properly licensed
• no real narrator was cloned
• no copyrighted music was used
• all ambient sounds are licensed or original
• the AI provider permits commercial use
• storage and streaming rights are documented
• an author approval record exists
==================================================
ADMIN AUDIO CONTROL
Create an admin area:
/admin/audio
Include:
• books
• chapters
• manuscript version
• audio version
• narrator profile
• pronunciation dictionary
• render status
• approval status
• file duration
• file size
• transcript
• chapter markers
• waveform data
• publish status
• replacement history
Statuses:
Draft
Generating
Under Review
Revision Needed
Approved
Published
Archived
Do not overwrite an approved audio file without version history.
==================================================
PRONUNCIATION DICTIONARY
Create editable entries for:
• character names
• aliases
• Cleveland locations
• streets
• restaurants
• trucking terminology
• Spanish words
• brands
• unusual slang
• family names
Each entry includes:
• written form
• approved pronunciation
• phonetic guide
• audio sample
• context
• approval status
Use the dictionary across every chapter.
==================================================
QUALITY REVIEW
Review every chapter for:
• incorrect pronunciation
• character voice drift
• pacing problems
• missing words
• added words
• repeated lines
• unnatural breaths
• robotic phrasing
• emotional mismatch
• distorted audio
• sudden volume changes
• awkward edits
• dialogue confusion
• incorrect dialect performance
• overacting
• underacting
• misread punctuation
• truncated sentences
• long dead air
• inconsistent chapter loudness
Compare the final audio to the manuscript word by word.
==================================================
FAILURE HANDLING
If audio fails to load:
Display:
The audio preview is having trouble loading.
You can still read the chapters now or try the player again.
Buttons:
Try Again
Read Chapter One
Contact Support
Never leave a blank player.
If progress saving fails:
Continue playback.
Do not interrupt the listener.
Attempt to save again later.
==================================================
PERFORMANCE
Use:
• streaming-friendly audio
• HTTP range requests
• preload metadata only
• lazy-load audio until requested
• CDN delivery
• cached waveform data
• compressed artwork
• persistent audio state
• background-safe playback where supported
Do not download both chapters automatically.
Do not block the book page while audio initializes.
==================================================
ANALYTICS
Track privacy-conscious events:
• player opened
• chapter started
• chapter paused
• chapter completed
• seek used
• speed changed
• transcript opened
• continued listening
• audio error
• signed-in progress synced
• Trucker J. chat opened
Do not record microphone audio without consent.
Do not send full transcripts of private voice conversations to general analytics.
==================================================
FINAL LAUNCH TEST
Before launch, verify:
• Chapter One uses the final manuscript
• Chapter Two uses the final manuscript
• narration is word-for-word accurate
• voice remains consistent
• narrator is original
• no real narrator is copied
• Cleveland names are pronounced correctly
• dialogue remains distinct
• Toya is not caricatured
• Latino characters are not caricatured
• slang sounds natural
• pacing changes with emotion
• pauses feel intentional
• audio is permanently rendered
• every playback is identical
• Play works
• Pause works
• Seek works
• rewind works
• forward works
• chapter switching works
• speed control works
• transcript sync works
• progress persists
• signed-in progress syncs
• mini player survives navigation
• mobile player works
• keyboard controls work
• screen readers work
• autoplay is disabled
• audio failure has a fallback
• author approval is recorded
• rights are documented
==================================================
FINAL EXPERIENCE
The listener should not feel as though a machine is reading text.
They should feel as though someone who understands Cleveland, trucking, family, pressure, loyalty, and long roads sat down and told them the story.
The performance should carry enough rhythm to feel alive.
Enough restraint to feel believable.
Enough emotion to make the family matter.
Enough tension to make the danger move closer.
Enough silence to let the listener feel what the characters cannot say.
The player should disappear into the experience.
The voice should become part of the LOADZ world.
By the end of Chapter Two, the listener should not be thinking about the technology.
LOADZ. Written by James Franks. Narrated... just for you.
James... was awakened by the noise... of a slamming truck door. As he sat up... he noticed he had missed several calls. Four from Toya... and one from Chi-Chi. Toya was his absolutely gorgeous wife... and the mother of his kids. Two years earlier... Chi-Chi was his direct pipe-line... to the Aztec Cartel.
I’ve been drivin' the same route for the past six months... which so happens to have a layover... in Kentucky. One day... when returning to my truck after takin' a shower... I was approached by a short Mexican named Chi-Chi... also known... as Taco. I was a little thrown off... by his approach to me. Being a trucker and all... red flags... went off immediately. Once I realized he posed no threat... I became more lax.
Chi-Chi asked me how long I had been a truck driver... tryin' to feel me up. I responded by saying... the past six years... then got straight to the point... "What's up though?"
Now... I’ve heard plenty of stories about how the cartels approached truckers... and attempted to manipulate... or flat-out threaten drivers... to transport drugs across the border into the states. Chi-Chi made sure he had my undivided attention as he spoke... askin' me how much I made a year... tryin' to speak English as best he could. "We’ve been watchin' you... for the past six months."
My mind switched into overdrive... as I began to process what was takin' place. Before I had a chance to respond... Chi-Chi indicated I could make my yearly salary... roughly... every ninety days. From the expression on my face... Chi-Chi had a gut feeling that he had chosen the right trucker... for the task at hand. Again... Chi-Chi never gave me a chance to reply. He handed me a cellphone... told me he would be in touch... then walked away.Chapter Two.
Still not quite sure what just took place... I climbed into my truck... starin' at the phone in my hand... tryin' to figure out why the Cartels chose me... out of all these damn truckers. I noticed I had nine missed calls... and five voicemails... from my wife, Toya... on my iPhone ten. I grabbed my phone... and called her.
She picked up on the fourth ring... soundin' like I had awakened her... out of her sleep. "Hey Sweetie... I see you called a few times. I was takin' a shower and the phone was in the truck. Sorry about that."
She answered... "Um hum... anyway... when are you comin' home? Me and the boys are missin' you. Shit... this pussy is missin' you."
I should be back tonight. I have a load to drop here in Kentucky... then I’m headed back to the City. "Tell daddy... how bad that pussy misses me. I know you've been playin' with it without me."
Toya burst out laughing. "Yeah... but it ain’t like the real thing. I need you."
"I know... I need you too... Mrs. L.B." That's a nickname I gave her. It stands for loose booty. I love the way that ass jiggles... when she walks.
"Well, hurry up... so I can put this loose booty... all on that dick!"
I laughed... "you crazy, bae."
"Yeah... crazy for you! Love you. See you tonight." "Love you too."
They should be reaching for the book.
BUILD THE LOADZ AUDIOBOOK PREVIEW EXPERIENCE
Create a premium audiobook feature for the first two chapters of LOADZ.
This is not a temporary text-to-speech demo.
Generate, approve, save, and securely host permanent audio files so every listener hears the exact same approved performance every time.
The audio must:
• resume from the same timestamp
• pause and restart without regenerating
• support seeking
• preserve the narrator’s approved performance
• work across desktop and mobile
• connect naturally to THE FOUNDATION AI
• allow visitors to hear Trucker J. guide them into the world of LOADZ
The goal is to make the opening chapters feel professionally narrated, culturally authentic, emotionally alive, and worthy of a commercial audiobook release.
==================================================
CREATIVE STANDARD
Create the emotional quality of a premium Audible-style audiobook without copying or impersonating any identifiable narrator, actor, celebrity, or real person.
Do not clone a real person’s voice.
Do not imitate a specific living narrator.
Instead, create an original fictional narrator voice with these qualities:
• Black American male
• approximately 38 to 48 years old
• raised in or deeply familiar with Cleveland, Ohio
• warm baritone
• grounded
• mature
• observant
• emotionally intelligent
• naturally street-aware
• confident without performing toughness
• capable of humor, tension, tenderness, and danger
• believable as someone who has driven long roads, raised a family, survived pressure, and learned when silence matters
The voice should carry the emotional energy associated with Trucker J.:
• provider
• protector
• husband
• father
• working man
• patient observer
• quiet authority
• somebody who thinks before he speaks
The narrator is inspired by Trucker J.’s emotional qualities.
The narrator must not claim to literally be Trucker J. unless the website clearly labels the performance as an in-character fictional reading.
==================================================
PUBLIC FEATURE NAME
Use:
HEAR THE STORY
Supporting line:
The road sounds different when somebody who knows the weight is telling it.
Alternative section title:
LISTEN TO LOADZ
Do not call it:
AI Voice Demo
Text-to-Speech
Generated Audio
Robot Narration
Audio Test
A small disclosure may appear beneath the player:
This preview uses an original AI-assisted performance created specifically for the LOADZ experience.
==================================================
CONTENT SCOPE
Create audiobook previews for:
LOADZ
Chapter One
Chapter Two
Use the final approved manuscript text exactly as written.
Do not:
• rewrite sentences
• clean up dialogue during narration
• change dialect
• replace slang
• add dialogue
• remove profanity
• summarize scenes
• alter character names
• censor language unless the author approves a separate clean version
• change story facts
• improvise beyond the manuscript
The audio should preserve the author’s wording while using performance, pacing, silence, and character distinction to bring it alive.
Before generating audio, confirm that the Chapter One and Chapter Two manuscripts are the approved final versions.
If multiple manuscript versions exist, stop and ask which version is canonical.
==================================================
NARRATOR VOICE PROFILE
Create one original narrator voice and preserve it as the permanent LOADZ narration identity.
VOICE CHARACTERISTICS
Pitch:
Low to medium-low baritone.
Not unnaturally deep.
Not a movie-trailer growl.
Tone:
Warm, textured, lived-in, controlled.
Cadence:
Natural Great Lakes and Black Midwestern rhythm.
The cadence may carry traces of Cleveland familiarity, but it must never become a caricature.
The narrator should sound like a real Black man from Cleveland telling a story to people who understand the city.
He should not sound like:
• a generic Southern narrator
• a New York stereotype
• an exaggerated “street” performance
• a preacher throughout the entire reading
• a radio announcer
• a commercial voice-over
• a rapper reading lyrics
• a motivational speaker
• an actor forcing slang
• an older man trying to sound young
• a young man trying to sound hardened
PRONUNCIATION
Respect:
• Cleveland neighborhood names
• local streets
• Black names
• Latino names
• trucking terminology
• vehicle names
• street slang
• regional phrasing
• character nicknames
• contractions
• intentionally nonstandard grammar
Create a pronunciation dictionary before production.
Every uncertain proper noun must be reviewed by the author.
Do not guess repeatedly across different chapters.
==================================================
NARRATION PERFORMANCE
The narrator should read prose and dialogue differently.
PROSE
• calm but cinematic
• clear without sounding overly formal
• measured enough to absorb detail
• slightly slower during reflection
• tighter during action
• never monotone
• never overacted
DIALOGUE
Give every character a distinct emotional posture without turning the reading into comedy or impersonation.
Differentiate characters through:
• tempo
• confidence
• breath
• sentence attack
• pause length
• emotional pressure
• vocal placement
Do not create extreme cartoon voices.
Do not use exaggerated accents.
Do not make women sound like parodies.
Use subtle performance changes.
==================================================
TRUCKER J. PERFORMANCE
Trucker J.’s dialogue should feel:
• calm
• masculine
• thoughtful
• responsible
• tired when the road has been long
• softer around Toya and the boys
• cautious around unfamiliar people
• firm when protecting his family
• internally conflicted when pressure begins building
He does not waste words.
He should sometimes pause before answering, as if calculating the weight of what he is about to say.
He should never sound ignorant.
He is a hardworking professional with street awareness, emotional intelligence, and real responsibilities.
==================================================
TOYA PERFORMANCE
Toya should sound:
• confident
• warm
• quick when joking
• direct when serious
• affectionate without becoming weak
• naturally beautiful in presence, not performed through a breathy voice
• grounded as a wife and mother
• capable of taking control of a room
Do not make Toya sound submissive, shrill, or overly dramatic.
==================================================
RICO PERFORMANCE
Rico should sound:
• charismatic
• forceful
• neighborhood-rooted
• confident
• capable of humor
• dangerous when disrespected
• respected without constantly yelling
His volume should not be his personality.
His certainty should carry the authority.
==================================================
GUNNER PERFORMANCE
Gunner should sound:
• lower-energy than Rico
• precise
• restrained
• observant
• emotionally guarded
• calculated
• capable of making a quiet sentence feel heavier than shouting
==================================================
CHI-CHI PERFORMANCE
Chi-Chi should sound:
• friendly
• measured
• intelligent
• bilingual or culturally fluent where the manuscript supports it
• professional
• more dangerous because of his calmness, not because of a forced accent
Do not exaggerate Mexican or Latino speech patterns.
==================================================
PACING SYSTEM
Use intentional pacing rather than one fixed reading speed.
GENERAL TARGET
Average narration:
Approximately 145 to 165 spoken words per minute.
Reflective passages:
Approximately 125 to 145 words per minute.
Normal dialogue:
Approximately 150 to 175 words per minute.
High-tension action:
Approximately 165 to 185 words per minute, while preserving clarity.
Never rush important names, emotional revelations, or scene-setting details.
==================================================
PAUSES
Use silence as part of the performance.
Suggested pause logic:
Comma:
Brief natural breath, not always a full stop.
Period:
Approximately 250 to 500 milliseconds depending on emotional weight.
Paragraph break:
Approximately 500 to 900 milliseconds.
Scene shift:
Approximately 1 to 1.8 seconds.
Major revelation:
Approximately 800 milliseconds to 1.5 seconds.
Chapter title:
Allow 1.5 to 2 seconds after the title before narration begins.
Do not apply these mechanically.
Adjust them to the emotional meaning of the sentence.
==================================================
EMOTIONAL DIRECTION
The narrator must understand the emotional function of each scene.
Before rendering, annotate the manuscript privately with:
• scene objective
• dominant emotion
• secondary emotion
• tension level
• character power balance
• relationship temperature
• pace
• required silence
• pronunciation notes
• transition notes
Suggested tension scale:
1:
Safe and domestic
2:
Curious
3:
Uneasy
4:
Suspicious
5:
Pressured
6:
Threatening
7:
Dangerous
8:
Immediate crisis
9:
Violent or emotionally explosive
10:
Major irreversible moment
Do not read every chapter at the same emotional intensity.
Contrast is essential.
The family scenes must feel warmer so the danger carries more weight.
==================================================
ENUNCIATION
The narrator should be clear without sounding polished beyond the world of the book.
Preserve natural speech features:
• contractions
• dropped endings where written
• rhythmic emphasis
• conversational compression
• authentic pauses
• local phrasing
• culturally natural stress
Do not pronounce every word as though reading a legal document.
Do not correct dialect through performance.
Do not over-pronounce slang.
The listener should understand every sentence while still feeling like the story belongs to its people.
==================================================
CLEVELAND SOUND AND CULTURAL TEXTURE
The reading should feel connected to Cleveland through:
• verbal rhythm
• emotional restraint
• blue-collar realism
• neighborhood familiarity
• weather awareness
• road and trucking references
• East Side and West Side distinctions where supported
• family-centered humor
• practical communication
Do not repeatedly say “Cleveland” to manufacture authenticity.
Authenticity should come from how the people speak, move, and respond.
==================================================
AUDIO PRODUCTION WORKFLOW
Do not generate the full chapters in one uncontrolled pass.
Use this production process:
Do not publish the first generated take automatically.
==================================================
PERMANENT AUDIO REQUIREMENT
The final approved audiobook preview must be pre-rendered and saved as permanent audio files.
Do not regenerate the audio every time a visitor presses Play.
This guarantees:
• identical performance
• identical pacing
• identical pronunciation
• consistent timestamps
• reliable pausing
• reliable resuming
• predictable cost
• consistent sound quality
• stable chapter markers
Create separate master files:
loadz-chapter-01-master.wav
loadz-chapter-02-master.wav
Create optimized streaming files:
loadz-chapter-01.mp3
loadz-chapter-02.mp3
Optional modern format:
loadz-chapter-01.m4a
loadz-chapter-02.m4a
Use WAV as the archival master.
Use MP3 or AAC/M4A for browser playback.
==================================================
AUDIO QUALITY
Master specifications:
• clean voice recording
• no clipping
• no digital distortion
• no obvious AI glitches
• consistent loudness
• controlled sibilance
• minimal room coloration
• no background hiss
• no abrupt cuts
• natural breaths preserved selectively
• breaths reduced only when distracting
• seamless edits
Suggested final loudness target:
Approximately -18 to -16 LUFS integrated for spoken-word playback.
Suggested true peak ceiling:
Approximately -1 dBTP.
Keep chapter loudness consistent.
Do not crush the voice with heavy compression.
==================================================
BACKGROUND SOUND
Narration should remain the priority.
Optional atmospheric sound may be used sparingly.
Possible intro texture:
• distant truck idle
• wet-road ambience
• soft Cleveland wind
• faint warehouse room tone
• page turn
Possible scene transitions:
• brief road noise
• distant city ambience
• soft household atmosphere
• restrained restaurant ambience
Rules:
• never place constant music beneath the entire chapter
• never overpower dialogue
• never use copyrighted commercial music
• never add gunshots or violent effects merely for excitement
• do not turn the audiobook into a radio drama unless separately approved
• allow listeners to disable enhanced ambience if two mixes are produced
Preferred initial release:
Narration-forward mix with only a short opening and closing sound signature.
==================================================
AUDIO PLAYER PLACEMENT
Add the player prominently on:
/books/loadz
Place it near:
Begin Reading Free
Use a paired choice:
READ THE FIRST TWO CHAPTERS
LISTEN TO THE FIRST TWO CHAPTERS
Also place a compact player inside the free-reading experience.
Suggested heading:
HEAR THE ROAD OPEN.
Supporting copy:
Listen to the first two chapters of LOADZ, performed in an original voice created for the world of the book.
==================================================
PLAYER DESIGN
Create a custom, accessible audiobook player.
REQUIRED CONTROLS
• Play
• Pause
• Restart
• Skip Back 15 Seconds
• Skip Forward 15 Seconds
• Previous Chapter
• Next Chapter
• Seek Bar
• Current Time
• Total Duration
• Playback Speed
• Volume
• Mute
• Transcript
• Chapter List
• Download only if approved
• Close or minimize where applicable
PLAYBACK SPEEDS
0.75×
0.9×
1×
1.1×
1.25×
1.5×
1.75×
2×
Default:
1×
Do not alter pitch when changing speed.
==================================================
PLAYER VISUAL DESIGN
The player should match the LOADZ website.
Use:
• charcoal
• deep black
• warm ivory
• restrained red accents
• brushed-metal or paper-inspired texture
• readable waveform or progress line
• clear typography
• large controls
• subtle motion
Do not use:
• generic browser-player styling
• neon waveform
• spinning records
• cassette imagery
• cartoon headphones
• excessive visualizers
• nonstop animation
Suggested layout:
LOADZ
CHAPTER ONE
[Chapter Title]
[15 sec back] [Play/Pause] [15 sec forward]
00:00 ━━━━━━━━━━━━━ 18:42
1× Chapters Transcript Volume
Below:
Narrated in an original performance created for the LOADZ experience.
==================================================
PROGRESS BAR
The progress bar must:
• show buffered audio
• show current playback position
• support click-to-seek
• support drag-to-seek
• support keyboard seeking
• show elapsed time
• show total time
• remain readable on mobile
• update smoothly
• not reset during ordinary page interactions
Use a larger invisible touch target around the visible bar.
==================================================
PLAYBACK PERSISTENCE
When a listener pauses:
Store the exact timestamp.
When they return:
Offer:
CONTINUE LISTENING
Chapter One
12:43 remaining
Resume from the saved position.
For logged-out users:
Store progress locally on that device.
For signed-in members:
Store progress securely in their account so it can continue across devices.
Store:
• book
• chapter
• timestamp
• total duration
• playback speed
• last listened date
• completion percentage
Do not reset progress unless the listener chooses Restart.
==================================================
MINI PLAYER
When audio continues while the visitor navigates within the website, display a compact persistent player.
Desktop:
Bottom edge or lower corner.
Mobile:
Bottom bar above safe areas.
Include:
• chapter name
• play or pause
• 15-second rewind
• progress
• expand player
• close
Do not restart the audio during internal page navigation.
Use a global audio state or persistent layout.
==================================================
CHAPTER SELECTION
Create a chapter list:
CHAPTER ONE
[Approved Chapter Title]
Duration
Play
CHAPTER TWO
[Approved Chapter Title]
Duration
Play
Show:
• current chapter
• completed chapter
• in-progress chapter
• locked future chapters where appropriate
Do not imply the full audiobook is available unless it exists.
Future chapters may show:
Full audiobook coming later
only if that release is genuinely planned.
==================================================
TRANSCRIPT SYNCHRONIZATION
Create an optional transcript view.
Requirements:
• exact manuscript text
• current paragraph highlighted
• click paragraph to seek
• auto-scroll optional
• user can disable auto-scroll
• accessible screen-reader structure
• no forced transcript display
• chapter and scene markers
• preserve punctuation
Do not expose private production annotations.
==================================================
ACCESSIBILITY
The player must support:
• keyboard controls
• visible focus
• screen readers
• clear labels
• large tap targets
• transcript
• high contrast
• reduced motion
• no sound autoplay
• no audio required to access the written chapters
• captions or transcript for atmospheric introduction where relevant
Suggested keyboard shortcuts:
Space:
Play or Pause
Left Arrow:
Back 5 seconds
Right Arrow:
Forward 5 seconds
Shift + Left Arrow:
Back 15 seconds
Shift + Right Arrow:
Forward 15 seconds
M:
Mute
T:
Open Transcript
Do not activate shortcuts while the user is typing in a form.
==================================================
AUTOPLAY
Do not autoplay chapter narration.
The visitor must press Play.
A muted visual invitation may appear.
Do not begin speaking when the page loads.
Do not surprise visitors with audio.
==================================================
THE FOUNDATION AI CONNECTION
Connect the audiobook experience to THE FOUNDATION.
On the LOADZ page, suggested AI actions may include:
• Read Chapter One to Me
• Continue My Audio
• What Did I Just Hear?
• Who Was Introduced?
• Explain This Without Spoilers
• Talk About Trucker J.
• Switch to Chapter Two
When the visitor asks:
“Read Chapter One to me.”
THE FOUNDATION should open the permanent audio player and begin playback only after clear user interaction.
Do not synthesize a new performance on demand when the approved chapter recording exists.
Use the saved master audio.
==================================================
TALKING TO TRUCKER J.
Create a clearly labeled optional fictional character experience:
TALK WITH TRUCKER J.
Supporting disclaimer:
A fictional AI experience grounded in the character and world of LOADZ.
This experience must remain separate from the narrator identity.
The audiobook narrator tells the story.
The Trucker J. conversation responds as an approved fictional interpretation of the character.
Do not blur them together without labeling.
Trucker J. conversational qualities:
• calm
• protective
• reflective
• family-focused
• practical
• measured
• road-wise
• unwilling to spoil the story
• capable of light humor
• aware that choices carry consequences
He may discuss:
• trucking life
• family
• being away from home
• responsibility
• Cleveland
• discipline
• early-story experiences
• themes in LOADZ
He must not reveal:
• hidden plot information
• future betrayals
• deaths
• major Book 2 outcomes
• information he would not know at that point in the story
Suggested opening:
TRUCKER J.:
Long road ahead.
You can listen to the first two chapters, ask me about life behind the wheel, or learn what it means to provide when everybody is counting on you.
Just know one thing.
I’m not giving away the whole route before you take the ride.
==================================================
TRUCKER J. VOICE EXPERIENCE
If voice conversation is enabled later, use an original voice consistent with the approved audiobook identity.
Do not clone a real person.
Do not imitate an identifiable narrator.
Clearly indicate that this is a fictional AI character experience.
Require user interaction before microphone access.
Provide:
• Start Voice Conversation
• Mute Microphone
• End Conversation
• Show Transcript
• Switch to Text
Do not record voice conversations without clear consent.
==================================================
AUDIO RIGHTS AND APPROVAL
Before publishing, confirm:
• James owns or controls the preview rights
• the final manuscript is approved
• the narrator voice is original or properly licensed
• no real narrator was cloned
• no copyrighted music was used
• all ambient sounds are licensed or original
• the AI provider permits commercial use
• storage and streaming rights are documented
• an author approval record exists
==================================================
ADMIN AUDIO CONTROL
Create an admin area:
/admin/audio
Include:
• books
• chapters
• manuscript version
• audio version
• narrator profile
• pronunciation dictionary
• render status
• approval status
• file duration
• file size
• transcript
• chapter markers
• waveform data
• publish status
• replacement history
Statuses:
Draft
Generating
Under Review
Revision Needed
Approved
Published
Archived
Do not overwrite an approved audio file without version history.
==================================================
PRONUNCIATION DICTIONARY
Create editable entries for:
• character names
• aliases
• Cleveland locations
• streets
• restaurants
• trucking terminology
• Spanish words
• brands
• unusual slang
• family names
Each entry includes:
• written form
• approved pronunciation
• phonetic guide
• audio sample
• context
• approval status
Use the dictionary across every chapter.
==================================================
QUALITY REVIEW
Review every chapter for:
• incorrect pronunciation
• character voice drift
• pacing problems
• missing words
• added words
• repeated lines
• unnatural breaths
• robotic phrasing
• emotional mismatch
• distorted audio
• sudden volume changes
• awkward edits
• dialogue confusion
• incorrect dialect performance
• overacting
• underacting
• misread punctuation
• truncated sentences
• long dead air
• inconsistent chapter loudness
Compare the final audio to the manuscript word by word.
==================================================
FAILURE HANDLING
If audio fails to load:
Display:
The audio preview is having trouble loading.
You can still read the chapters now or try the player again.
Buttons:
Try Again
Read Chapter One
Contact Support
Never leave a blank player.
If progress saving fails:
Continue playback.
Do not interrupt the listener.
Attempt to save again later.
==================================================
PERFORMANCE
Use:
• streaming-friendly audio
• HTTP range requests
• preload metadata only
• lazy-load audio until requested
• CDN delivery
• cached waveform data
• compressed artwork
• persistent audio state
• background-safe playback where supported
Do not download both chapters automatically.
Do not block the book page while audio initializes.
==================================================
ANALYTICS
Track privacy-conscious events:
• player opened
• chapter started
• chapter paused
• chapter completed
• seek used
• speed changed
• transcript opened
• continued listening
• audio error
• signed-in progress synced
• Trucker J. chat opened
Do not record microphone audio without consent.
Do not send full transcripts of private voice conversations to general analytics.
==================================================
FINAL LAUNCH TEST
Before launch, verify:
• Chapter One uses the final manuscript
• Chapter Two uses the final manuscript
• narration is word-for-word accurate
• voice remains consistent
• narrator is original
• no real narrator is copied
• Cleveland names are pronounced correctly
• dialogue remains distinct
• Toya is not caricatured
• Latino characters are not caricatured
• slang sounds natural
• pacing changes with emotion
• pauses feel intentional
• audio is permanently rendered
• every playback is identical
• Play works
• Pause works
• Seek works
• rewind works
• forward works
• chapter switching works
• speed control works
• transcript sync works
• progress persists
• signed-in progress syncs
• mini player survives navigation
• mobile player works
• keyboard controls work
• screen readers work
• autoplay is disabled
• audio failure has a fallback
• author approval is recorded
• rights are documented
==================================================
FINAL EXPERIENCE
The listener should not feel as though a machine is reading text.
They should feel as though someone who understands Cleveland, trucking, family, pressure, loyalty, and long roads sat down and told them the story.
The performance should carry enough rhythm to feel alive.
Enough restraint to feel believable.
Enough emotion to make the family matter.
Enough tension to make the danger move closer.
Enough silence to let the listener feel what the characters cannot say.
The player should disappear into the experience.
The voice should become part of the LOADZ world.
By the end of Chapter Two, the listener should not be thinking about the technology.
LOADZ. Written by James Franks. Narrated... just for you.
James... was awakened by the noise... of a slamming truck door. As he sat up... he noticed he had missed several calls. Four from Toya... and one from Chi-Chi. Toya was his absolutely gorgeous wife... and the mother of his kids. Two years earlier... Chi-Chi was his direct pipe-line... to the Aztec Cartel.
I’ve been drivin' the same route for the past six months... which so happens to have a layover... in Kentucky. One day... when returning to my truck after takin' a shower... I was approached by a short Mexican named Chi-Chi... also known... as Taco. I was a little thrown off... by his approach to me. Being a trucker and all... red flags... went off immediately. Once I realized he posed no threat... I became more lax.
Chi-Chi asked me how long I had been a truck driver... tryin' to feel me up. I responded by saying... the past six years... then got straight to the point... "What's up though?"
Now... I’ve heard plenty of stories about how the cartels approached truckers... and attempted to manipulate... or flat-out threaten drivers... to transport drugs across the border into the states. Chi-Chi made sure he had my undivided attention as he spoke... askin' me how much I made a year... tryin' to speak English as best he could. "We’ve been watchin' you... for the past six months."
My mind switched into overdrive... as I began to process what was takin' place. Before I had a chance to respond... Chi-Chi indicated I could make my yearly salary... roughly... every ninety days. From the expression on my face... Chi-Chi had a gut feeling that he had chosen the right trucker... for the task at hand. Again... Chi-Chi never gave me a chance to reply. He handed me a cellphone... told me he would be in touch... then walked away.Chapter Two.
Still not quite sure what just took place... I climbed into my truck... starin' at the phone in my hand... tryin' to figure out why the Cartels chose me... out of all these damn truckers. I noticed I had nine missed calls... and five voicemails... from my wife, Toya... on my iPhone ten. I grabbed my phone... and called her.
She picked up on the fourth ring... soundin' like I had awakened her... out of her sleep. "Hey Sweetie... I see you called a few times. I was takin' a shower and the phone was in the truck. Sorry about that."
She answered... "Um hum... anyway... when are you comin' home? Me and the boys are missin' you. Shit... this pussy is missin' you."
I should be back tonight. I have a load to drop here in Kentucky... then I’m headed back to the City. "Tell daddy... how bad that pussy misses me. I know you've been playin' with it without me."
Toya burst out laughing. "Yeah... but it ain’t like the real thing. I need you."
"I know... I need you too... Mrs. L.B." That's a nickname I gave her. It stands for loose booty. I love the way that ass jiggles... when she walks.
"Well, hurry up... so I can put this loose booty... all on that dick!"
I laughed... "you crazy, bae."
"Yeah... crazy for you! Love you. See you tonight." "Love you too."
They should be reaching for the book.