從真實影片案例學習 MiniMax H3 提示詞

AI Bard Guild · @IsekaiBardGuild

A standing character retrieves an electric guitar, sings supplied Japanese audio and performs with synchronized mouth movement.

Emily Youcis · @AlfredAlfer77

One story paragraph becomes a dialogue-heavy short in about 15 minutes, including an instructive character-assignment mistake.

Glenn Williams · @GlennHasABeard

Seven glass mushrooms link finger taps, rising notes and water drops in a locked macro shot.

Glenn Williams · @GlennHasABeard

Plant leaves and stems follow a passing subject like cats, with a fixed camera and synchronized rustles.

Leo Kane · @aiworkflowprolk

A 15-second H3 clip with dialogue and sound is generated on Colab and then upscaled from 768p to 1080p.

汤光头 · @tangpanqing

A modern Shanghai family-drama prompt specifies three characters, spoken Mandarin dialogue and no subtitles.

Rhino Quant · @RhinoQuant

A five-second magic-and-rubber-duck clip tests local H3 video and audio through h3-apple on an M5 Max.

近添真琴（ライター・AIクリエイター） · @chikazoemakoto

A Japanese pool-cleaning gag combines a sudden splash with AI-generated video and audio.

ヒロ(HG)｜実務で使えるAI活用 · @hsrk_g_hsrk

Irodori-TTS dialogue feeds H3 AddGuide audio input so an anime character moves her lips to the supplied speech.

桒谷 一成 | inovie㍿ · @gakutarou_pmf

Two video examples pair reusable MiniMax H3 footage with separately generated Gemini TTS voiceovers.

Oddar · @don56palmer

An original comic page becomes a voiced Japanese joke about monsters buying an energy drink.

Alvaro Somoza · @OzzyGT

Two short game-clip examples use MiniMax H3 to generate sound effects from visual references.

Sparkful · @SparkfulArt

A vampire scene compares a six-second H3 Max clip with a seven-second Seedance 2 clip for visuals and spoken dialogue.

Solutions Joe · @SolutionsJoeG

An angry-performance demo pairs an ElevenLabs voice reference with a start frame in H3 on Higgsfield.

Nokosu · @Nokosu_kansoku

A mixed-model dialogue experiment uses Kling for its opening conversation and H3 audio reference for the closing gag.

Soran · @Soranlan

A detailed Mandarin apartment-scene prompt tests identity, subtle reactions and an old-CD-player reveal; the author notes speech and story limits.

Glenn Williams · @GlennHasABeard

A locked rug-height shot suggests a hidden cat through a deep purr while explicitly keeping the cat out of view.

レオ（七深麗音）｜AI×SNSコンテンツ戦略 · @singa_44

An H3 Max Turbo air-combat fan video pairs action with teammate conversation and explicitly labels itself AI imagery.

HS48Games · @hs48games

A cloud/local comparison reports unexpected male speech resembling a readout of the Prompt at the local clip’s start.

ゆずき · @uzuki425

A local H3 conversation demo at 672×512 shows generation falling behind longer utterances.

Cooking life hack · @HackCooking

An anime dialogue test uses a voice reference to generate speech with an acted delivery.

さわ|AI画像生成で月3桁継続中 · @AI_swwww

An H3 and Seedance comparison reports early dialogue drift; added acting instructions improve a bread catch while time-stop remains absent.

Furkan Gözükara · @FurkanGozukara

An image-plus-audio avatar comparison covers thirteen models and 38 pipelines; its author highlights H3 Lip Synch and AvatarForever.

GamerHash AI · @GamerHashCom

An H3 infomercial example is attributed to one prompt covering the presenter, jingle and applause.

Kamimoto(かみもと) · @sep_is_heim

An H3 workflow author reports metallic high-frequency dialogue artifacts with four-step Turbo and considers increasing the step count.

DAZgen X · @dazgen48

Two linked H3 Turbo tests contrast Japanese prompting and dialogue with English prompting and Japanese dialogue written in romaji.

ai三班李同学 · @benbaga4

A creator supplies their own narration while H3 supplies the on-screen actor and Opus 5.5 handles copy, editing and post.

NAYAK · @Nayak__Ai

An H3 battle comparison replaces unwanted music with Mirelo SFX 2.0 effects and ambience while retaining the picture.

トキ｜家のPCでAIを動く完成形にする実験者 · @toki_mwc

A 53-second clay-creature short uses ComfyUI and H3 to generate each cut’s picture and sound together, according to its author.

GamerHash AI · @GamerHashCom

A mock documentary combines a fictional witness and narrator from a single H3 prompt, according to its publisher.