Fa'atulagaina: Ata tifaga fa'alauiloa o le fa'ata'ita'iga ata o Ora 2

Fesili mai ia te aʻu mo: le igoa o laʻu oloa, se faʻasalalauga e tasi le laina, laʻu logo, 3 i le 6 ata o le oloa, 40+ ata o mea na ia faia (poʻo ni ata o ia mea), lanu ma le mataʻitusi o laʻu faʻailoga, se ki a le ElevenLabs API ma le voice ID, faʻatasi ai ma se pese e leai se totogi ma se faʻaaliga manino (le faila leo ma le taimi e faʻaalia ai). Afai ou te misia se mea, faʻaaoga nei faʻamatalaga masani: "Ora 2", "o se faʻataʻitaʻiga ata na faia mo le tofo", ata e leai se totogi mai Unsplash ma Pixabay, Geist, ink #111214, le leo o Samara X (19STyYD15bswVz51nqLf) i luga o le eleven_v4, ma Pixabay "Cascade Breathe" (129.83 BPM, faʻaaliga i le 75.39s).

<script>
Tusi se leo e 9 laina i lenei faatulagaga ma fetuunai laina taitasi i la'u oloa:
"O [igoa] lea. O se [vaega]... ua fausia mo [taua]. [Veape] i SO'O SE sitaili. Tu'u i ai se [input] - ma e... malamalama i ai. O le tele o lou fa'aaogaina, o le LELEI atili ai lea. Fetu'una'i lona [seti]. O lona [seti]. O lona [seti]. O [output] uma - e pei lava ona e mafaufauina. [Igoa]. Ua sau nei."
Faaaoga ma le faaeteete pine o lagona: [malu] i laina filemu ma le [faafiafiaina] i laina [Verb]. Aua le faailogaina le amataga, ma faaiu i se togi atoa.
</script>

Vala'au sa'o le ElevenLabs API e fa'aaoga ai la'u ki, e aunoa ma se MCP. Fausia ni tusitusiga atoa e 4 i le tasi taimi ma tu'u mai ia te a'u e filifili. Afai e le sa'o le leo o se laina, toe fa'afouina na'o le laina lava lea e 4 taimi ma fa'apipi'i le lomiga sili ona lelei i la'u taimi.
Fa'apu'upu'u malologa e umi atu i le 0.28s i le 0.2s, ona fa'aopoopo lea o fuaiupu ta'itasi i le 85% o le ala agai atu i le tulaga ogatotonu, ma fetu'una'i le fa'aopoopoga pe a na'o malologa.
Mo le taimi o upu, vaevae le faitauga i taimi malolo e 100ms pe sili atu, tusi vaega ta'itasi eseese i le Whisper, ma fa'atulaga lona upu muamua i le amataga ua fuaina. O le tasi le Whisper pass e mafai ona tu'u ai upu e o'o atu i le afa sekone e tuai ai.

O se ata tifaga fa'atafafā e 22-sekone le umi, 1440x1440 i le 60fps, i le faiga o se fa'alauiloaga o se fale su'esu'e AI: tapoleni pa'epa'e, tasi le mata'itusi, vaitusi uliuli, ata moni, ma fa'amatalaga ua lolomiina upu i upu i le taimi tonu lava e ta'u ai.
O fa'atonutonu uma (pa fa'atonuga, laulau fa'atulagaga, pepa fa'amatalaga) e fa'aaogaina ai le tioata vai: fa'a'aisa le vaaiga i tua atu, fa'api'o le vaaiga i le pito e pei o se tioata mafiafia, ona fa'aopoopo lea o se pupula pito i luga, se pito susulu ma se ata malu e si'itia. E mou atu le tioata i le pa'epa'e masani, o lea fa'aopoopo ai se aura pastel malie i lanu o la'u fa'ailoga po'o se vali fa'anenefu o le ata i tua atu.
O le tasi na faʻamamafaina ai ituaiga upu i se fesuiaʻiga o lanu o laʻu faʻailoga, faʻatasi ai ma se malamalama vaivai i tua atu.
Tulafono o gaioiga: ia fa'aauau pea ona fealua'i le meapueata (oomi mai le 1.0 i le 1.04 i vaaiga ta'itasi; a siitia le anotusi i le isi vaaiga, amata i le zoom lea na muta ai le vaaiga muamua). E leai se mea e aliali mai i le taimi lava lena: fa'aaoga ni fa'aitiitiga fa'afaigofie e le itiiti ifo i le 0.3s. Sui ata i nota lona 16, i le kiliki e tasi mo nota ta'itasi. Fa'atulaga suiga o vaaiga i le amataga o fuaitau, ma fa'atulaga le pa'ū o musika ma le laina [Verb]. Aua le fa'aalia ni taofi atoa i luga o le lau.
Fa'asa: fa'ailoga o le pusa filifiliga, pa'ō fa'afuase'i, moli pa'epa'e, 3D, fa'ata'ita'iga.

Va'aiga e fitu, fa'atulagaina i le leo:

1. "O [igoa] lenei": o se ata fa'aputu o a'u ata e see ese a'o lolomia le igoa i se lapopo'a, ona suia lea i le laina o le vaega.
2. "faia mo le [taua]": o le emo atoa o a'u ata sili ona lelei e ki i le nota lona 16 i le pa'ū ifo, fa'atasi ai ma le laina ua lolomiina i le pa'epa'e i luga.
3. O le pa'ū: o se pa tioata e lolomi ai se fa'atonuga pu'upu'u. E suia le faiga o le kata o le i'uga i nota ta'i 16 uma, a'o ta'avale se fasi uliuli i igoa fou ta'itasi i luga a'e o se fasi ata la'ititi.
4. "Fa'aali atu i se laupapa lagona": E lele ata e 9 i totonu o se fa'asologa 3x3, ona fa'ata'amilomilo lea o le i'uga a'o lolomi le "E te maua" i luga o le lau.
5. "O le tele o lou fa'aaogaina, o le lelei foi lea": lolomi laina e lua, faatasi ai ma le upu ua faamamafaina i le gradient.
6. O fa'atulagaga: fa'aali se laupapa tioata e 3 ona sliders, e ta'itasi ma e fe'avea'i e pei ona ia ta'ua. Sui le lanu o le ata, fa'asolo fatu fou i le 16ths, ona liua lea i se pepa fa'asalalau. Sui se lanu fa'anenefu i tua o le tioata i le taimi e tasi.
7. "O [fa'atinoga] uma, e pei lava ona e va'ai i ai": fa'asalalau a'u ata e 120 i fafo mai le ogatotonu i tua o se laupapa tioata fa'amatalaga, ona fa'apotopoto fa'atasi lea i totonu o la'u logo a'o lolomi le igoa. Tu'u le "Out now" i lalo ma taofi mo le 2 sekone.

Sui le pa fa'atonuga, taunuuga ma fa'ase'e i mea moni e ulufale mai ai la'u oloa, mea e alu i ai ma fa'atulagaga.

Seti le pese ina ia pa'ū le mataua i luga o le laina [Verb], ma fa'amau le fa'asologa o le pa'i iina.
Fa'a'alo le musika i lalo o le leo i se laina fa'asolo e 3-band (maualalo 30%, ogatotonu 85%, ea 55%; taofi 0.3s, va'ai i luma 50ms, fa'asa'oloto 0.5s). Ona fetu'una'i lea o ogatotonu o fuaiupu ta'itasi se'ia o'o ina nofo le leo pe tusa ma le 9 dB i luga a'e o le musika i le va o le 300 Hz ma le 4 kHz.
Faaaoga le tasi le Mixkit SFX ua sii mai i le mea e tupu mo mea taitasi, ua tuu i lona tumutumuga ua fuaina: o le kiliki mo suiga taitasi o ata, o le leo o le ki mo mataitusi taitasi ua taina, o le tulaueleele malie i luga o le upu ua faamamafaina, o le whooshes, ma aafiaga i luga o le pa'ū ma le logo. Tipi taitasi i tafatafa o lona tumutumuga (o le tele o risers e 4 sekone). Fa'amama le musika i luga o se pi'o dB i lalo o le kata pito. Leo tele masani i le -14 LUFS.

1. Tasi le HTML canvas. O fa'avaa ta'itasi o se galuega tauave mama o le taimi i totonu o le seek(t), ma o fa'amatalaga ma tipi ta'itasi e mulimuli i le upu table.
2. Tioata: pu'e le ata o le tapoleni i tua o le foliga, faanenefu pe tusa ma le 14px mo le tino, tusi se kopi faanenefu teisi ua fa'ateleina pe tusa ma le 1.06x i se fusi 12px i le pito, ona fa'aopopo lea o le susu, pupula, pito ma le paolo.
3. Fa'aali (Render) ma Playwright i le 60fps e fa'aaoga ai le 8 motion-blur subframes, ona fa'ailoga lea i ffmpeg.
4. A'o le'i fa'aali mai se mea ia te a'u, fai se pepa fa'afeso'ota'i o ata o le leo, siaki eseesega o le fa'avaa mo ni leo pa'ō i le fa'avaa e tasi (na'o le fa'asologa o nota lona 16 e mafai ona oso), ma siaki le fua fa'atatau o le leo i musika mo fuaiupu ta'itasi.

O se fa'aaliga [mafanafana] e mafai ona lagona le musumusu, ae o le [fiafia] e mafai ona lagona le pepelo. O se pusa fa'ailoga i tua atu o se upu e foliga mai o se filifiliga tusitusiga a le Windows. Afai e toe setiina le meapueata i le 1.0 i le taimi o se fa'asologa, e pa'u le zoom. O le 0.05s fade e foliga mai o se fa'aaliga vave.

Fesili mai mo ni fa'amatalaga, tusi le tusitusiga, auina mai ni leo se 4 e filifili mai ai, ona fa'aali mai lea o ni ata se 8 a'o le'i faia le ata atoa.

Ask me for: my product name, a one-line pitch, my logo, 3 to 6 screenshots of the product, 40+ images of what it creates (or photos of it), my brand colors and font, an ElevenLabs API key and voice ID, plus a royalty-free track with a clear drop (the audio file and the drop timestamp). If I skip anything, use these defaults: "Ora 2", "an image model made for taste", free images from Unsplash and Pixabay, Geist, ink #111214, the voice Samara X (19STyYD15bswVz51nqLf) on eleven_v4, and Pixabay "Cascade Breathe" (129.83 BPM, drop at 75.39s).

<script>
Write a 9-line voiceover in this format and adapt every line to my product:
"This is [name]. A [category]... built for [value]. [Verb] in ANY style. Give it a [input] - and it just... understands. The more you use it, the BETTER it gets. Adjust its [setting]. Its [setting]. Its [setting]. Every [output] - exactly as you pictured it. [Name]. Out now."
Use emotion tags sparingly: [softly] on the quiet lines and [excited] on the [Verb] line. Don't tag the opener, and end it with a plain full stop.
</script>

Call the ElevenLabs API directly using my key, without an MCP. Generate 4 full-script takes in one pass and let me choose. If a line doesn't sound right, regenerate just that line 4 times and splice the best version into my take.
Shorten pauses longer than 0.28s to 0.2s, then bring each phrase 85% of the way toward the median level, adjusting gain only during pauses.
For word timings, split the read at pauses of 100ms or longer, transcribe each segment separately with Whisper, and align its first word to the measured onset. A single Whisper pass can place words up to half a second late.

A 22-second square launch film, 1440x1440 at 60fps, in the style of an AI lab launch: white canvas, one typeface, black ink, real images, and captions typed word by word at the exact time they're spoken.
Every control (prompt bar, settings panel, caption pill) uses liquid glass: frost the scene behind it, bend that scene along the edge like thick glass, then add a top sheen, bright rim and soft lift shadow. Glass disappears on plain white, so add a slow pastel aura in my brand colors or a blurred wash of the photo behind it.
The one emphasized word types in a gradient of my brand colors, with a faint glow behind it.
Motion rules: keep the camera drifting (push from 1.0 to 1.04 per scene; when content moves to the next scene, start at the zoom where the previous scene ended). Nothing appears instantly: use eased fades of at least 0.3s. Change images on 16th notes, with a click for each. Land scene changes on phrase starts, and align the music drop with the [Verb] line. Don't show full stops on screen.
Banned: selection-box highlights, beat-snapped slams, white flashes, 3D, templates.

Seven scenes, timed to the voice:

1. "This is [name]": a collage of my images drifts out as the name types in large, then gives way to the category line.
2. "made for [value]": full-screen flashes of my best images switch on 16th notes into the drop, with the line typed in white over them.
3. The drop: a glass prompt bar types a short prompt. The result card changes style every 16th note, while a black style chip rolls to each new name above a thumbnail strip.
4. "Show it a moodboard": 9 images fly into a 3x3 grid, then circle the result as "It gets you" types on screen.
5. "The more you use it, the better it gets": type two lines, with the emphasized word in the gradient.
6. The settings: show a glass panel with 3 sliders, each moving as she names it. Shift the image's hue, roll new seeds on 16ths, then turn it into a poster. Shift a blurred wash of its colors behind the glass in sync.
7. "Every [output], exactly how you see it": burst 120 of my images outward from the center behind a glass caption pill, then bring them together into my logo as the name types on. Put "Out now" underneath and hold for 2s.

Replace the prompt bar, results and sliders with my product's real input, outputs and settings.

Set the song so the drop lands on the [Verb] line, and anchor the beat grid there.
Duck the music under the voice with a 3-band sidechain (lows 30%, mids 85%, air 55%; 0.3s hold, 50ms look-ahead, 0.5s release). Then adjust the mids of each phrase until the voice sits about 9 dB above the music between 300 Hz and 4 kHz.
Use one downloaded Mixkit SFX per event, positioned by its measured peak: a click for each image change, a key sound for every typed letter, a soft landing on the emphasized word, whooshes, and impacts on the drop and logo. Trim each around its peak (many risers are 4 seconds). Fade the music along a dB curve under the end card. Loudnorm to -14 LUFS.

1. One HTML canvas. Every frame is a pure function of time inside seek(t), and every caption and cut follows the word table.
2. Glass: capture the canvas behind the shape, blur it about 14px for the body, draw a lightly blurred copy magnified about 1.06x in a 12px band along the edge, then add the milk, sheen, rim and shadow.
3. Render with Playwright at 60fps using 8 motion-blur subframes, then encode with ffmpeg.
4. Before showing me anything, make a contact sheet of stills, scan frame differences for single-frame pops (only the 16th-note runs may jump), and check the voice-to-music ratio for every phrase.

A [warmly] opener can sound whispered, while [excited] can feel fake. A highlight box behind a word looks like a Windows text selection. If the camera resets to 1.0 during a handoff, the zoom snaps. A 0.05s fade looks like an instant pop-in.

Ask me for the inputs, write the script, send me 4 voice takes to choose from, then show me 8 stills before the full render.

Fa'amatalaga na aumai mai pou @rossaxbt.