Template centerUpdated Oct 30, 2026

Kinetic Tech · Talking-Head Explainer

Metallic type and spatial motion. Chinese landscape/portrait; English landscape and portrait basic expressions B1–B4, with local face centering.

Full template

A cohesive set of composable animation and motion components for creating an entire video around your content.

PREVIEWdefault1080×1920 · 60fps

View creator profile · @adunext

Get and use this template

FreeMIT Open License

View license details

The template code is MIT-licensed: you may use it commercially, modify and redistribute it, keeping the copyright and license notice. Brands, likenesses, fonts and third-party assets follow the licenses included in the package.

Free signup. Free template download.

One account connects AI asset matching, music and more creative services.

View details · uses, materials and requirements

Before you download

Suitable for: 用强节奏金属字、焊花、机械翻牌与空间卡片讲清观点和关系
Prepare these inputs: What do you want to talk about? Just send me any copy or materials you already have. · If the topic is already given, draft short copy for review without asking again. · Ask only this one question when there is no topic.
Keep the presenter footage already provided. · With footage, bind its actual size and face position; text-only content uses B1–B4. · If no presenter is requested, don't ask; only ask for missing presenter footage when explicitly requested.
Keep existing captions or measured speech timing. · Captions follow the real voice; text animations without voice need no extra caption file. · No need to fabricate word timings; if captions aren't requested, don't ask.
Use the user's existing screen recordings or work. · Read count, aspect ratio and duration requirements only for the selected shots. · Don't ask if no media shots are selected; if selected and assets are missing, ask only for those assets.
Keep the narration already provided or explicitly requested. · Use existing audio at its real duration, without changing speed. · If no narration is requested, proceed directly; if requested but the file is missing, ask only for the needed recording, without explaining capabilities.
Keep the background music already provided or explicitly specified. · Bind the actual track and known source info; keep the native action sound effects. · If no music is requested, proceed directly; if requested but the file is missing, ask only for that audio, without a mandatory "with/without music" question.
Production mode · Always use this episode's user materials
Language · Chinese or English; visuals, audio and captions are localized separately
Aspect ratio · Chinese and English: landscape1920×1080 at60fps. Portrait1080×1920 supports the declared B1–B4 basic-expression groups in both languages; English native source groups remain landscape only. Portrait film/chapter titles fit18 characters and presenter labels10, with actual measured line layout.
Project title · From this video's user content, not the original topic; in portrait, project and chapter titles are max 18 characters each, with the full text in captions
Presenter label · Use the name the user goes by; may be empty if they explicitly won't appear on camera; portrait max 10 characters
Shot groups · Choose groups by content first; the assistant then fills inputs from the catalog and schema. Users don't fill in parameter sheets.
On-screen content · Fill in per the slots and capacity in core/content.schema.json, keeping the user's original meaning
Action sounds · native keeps the corresponding native sound effects, none turns them off explicitly; narration and music are chosen separately
Captions · Only trailing commas or periods are removed from display; internal punctuation and actual word intervals are kept.
Flexible expression · B1 keyword and explanation, B2 one to five neutral points, B3 two perspectives and a conclusion, B4 summary and action. Preserve the original meaning; paragraphs use explicit one or two lines; extend only within declared quiet reading windows; native actions stay at original speed.

Run locally with · Runtime verification

Requires file and command access plus the declared runtime dependencies.

Version and license: 0.2.44-alpha.5 · licenses/TEMPLATE-LICENSE.txt
Run locally with: Requires file and command access plus the declared runtime dependencies.
kinetic-user-core/1 · kinetic_bootstrap.py · motion.lock.json
Runtime verification: Not verified

Summary from the template introduction (README.md) and this release’s motion.json, params.schema.json and guide.json. Missing details remain unverified.

How do I use the download?

Download .motion and keep the original.
Give it to an AI assistant with file and command access.
Check runtime dependencies, preview, then export.

Wait for the attachment upload to finish, then ask the assistant to confirm it can read the file. If it keeps spinning, retry or use the ZIP compatibility download. A .motion file is already a ZIP container; do not compress it again. If it remains unreadable, place the original in a local project folder the assistant can access. Read the file handoff guide →

Package details and handoff text

file: 385712-0.2.44-alpha.5.motion
size: 26 MB
sha256: 914e22378bd3…

Give it to your AI assistant

Use this MotionGit template for my content. First actually read the attachment or original package in this project: confirm its filename, size and read-only outer entry list. If upload is pending or access fails, say so; do not claim receipt or suggest wrapping .motion in another ZIP. Keep the original. Read 00-START-HERE.md and assistant/guide.json first when present, verify the bundled entry and identity, then read the actual material contract. Do not unpack all contents, download another Skill/host or copy an identifier. Without the bundled entry, read the actual motion.json, guide.json and README. Explain editable content, required assets and runtime first. Do not run unreviewed code or display a Key. Continue according to my content and authorization.

Instructions for AI →

Your download includesyour personal API key. Keep .motion files containing your personal key private. Share the template link; visits do not earn credits. Referral rewards require an eligible friend’s actual payment.

Runtime details

Runtime: kinetic-user-core/1
Output: 1080×1920
Frame rate: 60 fps
Duration: 01:05
Params: 26
Styles: 1
Network: offline
Spec: 0.1.0-draft.1

About

Narration Technology metallic texture 3D Typography Landscape and Portrait Face centered

Narrated videos
2.7K downloads
Updated 2 days ago
Authoring tool codexDeclared by creator

Static checks

Package structure and path safetyRoot manifest: motion.jsonParameters and style branchesUser guidance and duration adaptationFile integrityPersonal information and credentialsInstructions and licensesAuthoring provenance

Passed: package structure and automated scans. These checks do not verify copyright ownership or guarantee safe code.

Free template

Free download. Open source. Make it yours.

Free to download, with no top-up required. Reuse and customize it: replace text, images and audio, and adjust visuals and animation.

Free assets and musicImage, video and sticker search plus 1,000+ music tracks your AI assistant can match.Asset guide →
Optional generationImage, voice and video generation are optional and billed separately.Creator API →
First time?Follow the tutorial from template to finished video.Open tutorial →

Template code follows the license on this page; fonts, images and music keep their own licenses. Free asset services have a daily allowance.

About this template

Metallic 3D typography, welding sparks, mechanical flip cards and spatial cards are suitable for expressing opinions through your own spoken words, photos, screen recordings and copywriting. Contains 23 native units and 4 sets of basic expressions from the current library, retaining native actions and independent sound effects.

This version supports

Chinese: 1920×1080 horizontal screen or 1080×1920 vertical screen, both 60fps.
English: 1920×1080 horizontal screen; newly added 1080×1920 vertical screen basic expressions B1–B4, both 60fps. The English-native S series still uses horizontal screen.
The Chinese Demo retains the reviewed vertical screen clip; the English Demo is updated to a 36-second vertical screen B1–B4 example, including native sound effects and CC0 soundtrack, without narration. The website proportion annotation corresponds to the default Chinese Demo.
Person pictures are recognized once, and up to five frames of the video are extracted and cropped according to the approximate center of the face. When not detected, it is normally centered and the detection result is reported.
Circular and rectangular character windows, top chapters, subtitles and media are typeset separately; the time in the upper right corner and the official circular .motion logo alternately fade in and out; 1080 wide vertical screen leaves 96px on the right and 88px on the top, and the preview and export are consistent.

Provide the assistant with the topic, copywriting and existing materials of this video, read the first reading instructions in the package, and follow the normal plan, prepare, preview, and render entrances to create it. After updating, create another instance and keep the old package and original project. Vertical items and chapter titles can each have a maximum of 18 characters, and character identifiers can have a maximum of 10 characters.

Licensing and verification scope

MIT license; included fonts and third-party parts follow their respective OFL/MIT licenses. User characters, narration, music and private materials are not included in the general package. The Chinese vertical screen and new material processes have been verified on macOS; this version is still a public beta version, and the actual rendering of English vertical screens B1–B4 has been completed; the English native S series vertical screen, Windows/Linux real machine, and all extreme capacities have not yet been completed.

English: Chinese supports landscape and portrait. English supports landscape plus portrait basic expressions B1–B4 at1080×1920/60fps. Native English S-series scenes remain landscape. The new36-second English portrait demo uses native sound effects and CC0 music, without narration. Updates retain your original Key, old package and projects.

Open source warehouse

https://github.com/motiontemplates/385712

Developer resources

Files Fork · 0

What you can replace

Parameter	Description	Type	Constraints	Default
`mode`Required	—	—	—	`—`
`language`Required	—	enum	zh / en	`—`
`layout`Required	Chinese supports landscape or portrait. English portrait supports B1–B4; English landscape retains all current groups.	enum	landscape / portrait	`—`
`duration`Required	—	number	≥ 4.3，≤ 600	`—`
`title`Required	—	string	≤ 45 characters	`—`
`presenterLabel`Required	—	string	≤ 45 characters	`—`
`scenes`Required	—	array	—	`—`
`content`Required	—	object	—	`—`
`media`Required	Global presenter, voice and music bindings are always explicit. Additional scene-specific visual roles retain their existing catalog.mediaPorts contracts. Read MEDIA.md for official music-prepare result mapping.	object	—	`—`
`soundEffects`Required	—	enum	native / none	`—`
`captions`Required	—	array	—	`—`
`basic`	—	—	—	`—`
`markWindows`	—	array	—	`—`
`narrationMap`	Complete shot ownership. Sentence beats use complete owned cue text and unchanged cue intervals. measured-word requires actual narrationTimingEvidence and wordIndices; see NARRATION-MAP.md.	array	—	`—`
`narrationTimingEvidence`	Real word ASR/forced-align result or sample-verified segmented voice source for measured-word beats, captions and component cues. See NARRATION-MAP.md; provenance is not semantic acceptance.	object	—	`—`
`captionTimingEvidence`	Actual listened-to complete SRT or local ASR/alignment source; provided-captions is not proof. Read NARRATION-MAP.md.	object	—	`—`
`narrationVisualMode`	Optional measured-sentence reveal for B1–B4. Omitted keeps the approved original choreography; no guessed word clocks.	string	original / follow-speech	`original`
`narrationLayout`	Optional new-production portrait centering for no-presenter B1/B2/B4. Omitted retains approved placement and all source clocks.	enum	original / balanced	`original`
`creationProfile`	/1 uses B1-B4; opt-in /2 compiles B and S03/S08/S10/S13 with native source clocks and real caption boundaries. Omit to retain old input. /kinetic-components/1 adds B1–B4 plus opt-in component scenes inside scenes[].	enum	narrated-story/1 / narrated-story/2 / kinetic-components/1	`—`
`requiredMediaRoles`	—	array	—	`—`
`header`	—	object	—	`—`
`captionPosition`	Optional portrait publish-preview ratios, chosen by the assistant after viewing the intended interface. Defaults bottom .24, left .08, right .16. Editorial preset, not an official device/platform guarantee.	object	—	`—`
`captionSafeArea`	Portrait uses an editorial bottom/right UI reserve. frame retains previous positioning for explicit legacy replay. Landscape defaults are unchanged.	enum	short-video / frame	`short-video`
`portraitComposition`	Portrait keeps the first published portrait choreography, with local presenter/caption safety adjustments. Historical values remain readable. Landscape is unchanged.	enum	presenter-caption/2 / legacy	`presenter-caption/2`
`captionLineBreak`	Optional Chinese caption auto-wrap that keeps numbers and measure words together. Omit for previous line breaks.	—	—	`—`
`sfxVoiceDucking`	Optional 50 ms RMS voice-priority native SFX ducking during spoken captions. Omit for previous audio.	—	—	`—`

Update history

Improvements, fixes and remaining issues in each release.

Update this template with your AI assistant

Give the old template to the assistant and send the following sentence. The new version is saved as a separate file, and the old packages and projects are retained.

Please press https://motiongit.com/update.md to update this MotionGit template directly, keep the old package and existing project, and tell me which new file to use when you're done.

v0.2.44-alpha.5LatestOct 30, 2026
No release notes have been provided for this version.
v0.2.44-alpha.4Oct 28, 2026
Finished videos made with the new version no longer show the small period title at the top; the main visuals and caption layout are unchanged.
View original text
新版制作的成片不再显示顶部周期小标题；画面主体和字幕布局保持不变。
v0.2.44-alpha.3Oct 28, 2026
When making a new talking-head video, you can choose the “Reveal by sentence” arrangement so that basic key-point cards highlight items one by one with the matching sentences, instead of the whole card staying unchanged for long. Existing plans can still use the original pacing.
View original text
制作新口播时，可选择“随句展开”编排，让基础要点卡按对应句子逐项突出，减少整张卡长时间不变。已有方案仍可使用原版节奏。

Show earlier updates (22)

v0.2.44-alpha.2Oct 27, 2026
Fixes the five optional explainer components possibly not being found by the production assistant after an update, and adds a fill-in example for the “Negation strike-through” component. The portrait presenter, captions and landscape visuals stay as in the previous version.
View original text
修复更新后五个可选讲解组件可能无法被制作助手发现的问题，并补充“否定划线”组件的填写示例。竖屏人物、字幕和横屏画面沿用上一版。
v0.2.44-alpha.1Oct 27, 2026
Adds five explainer components: memory panel, path straightening, many-to-one, one-to-one mapping and next-episode preview. In portrait videos the presenter is now placed below the content, centered horizontally, with a large card, card, round window or small corner circle chosen automatically for each section and smooth transitions between sizes. Portrait captions move lower and are centered, avoiding the platform's caption area at the bottom. Landscape visuals are unchanged.
View original text
新增五个讲解组件：记忆面板、路径拉直、多点聚一、一对一映射、下期预告。竖屏人物统一放在内容下方、水平居中，按每段内容自动选择大卡、卡片、圆窗或角落小圆窗，档位之间平滑过渡；竖屏字幕下移并左右居中，避开平台底部文案区。横屏画面不变。
v0.2.40-alpha.1Oct 27, 2026
The portrait presenter regains this template's original camera moves and shape changes, with position adjusted locally only where it might cover content. Captions are aligned to the center of the frame and leave room for the platform's text at the bottom. The extra period title beside the logo is removed; the time and round badge are kept.
View original text
竖屏人物恢复本模板原有的镜头移动与形状变化，仅在可能遮挡时局部调整位置。字幕按画面正中对齐，并为平台底部文案留出空间。移除 Logo 旁多余的周期标题，保留时间与圆形标识。
v0.2.38-open-betaOct 27, 2026
Fixes preview or export of an already prepared project possibly failing when using the new production assistant. The template's visuals, captions and sound design are unchanged; existing layouts and the demo remain available.
View original text
修复使用新版制作助手时，预览或导出已准备工程可能失败的问题。模板画面、字幕与声音设计保持不变；原有版式和 Demo 继续可用。
v0.2.37-open-betaOct 26, 2026
Adds optional components for progress, trends, steps and more; existing layouts remain available. Fixes the portrait presenter drifting to the bottom-right corner, the duplicate title next to the timer, and line breaks in native captions. The existing demo is unchanged.
View original text
新增可选的进度、趋势、步骤等表达组件，原有版式继续可用。修复竖屏人物偏到右下角、计时旁重复标题及原生字幕换行的问题。原有 Demo 保持不变。
v0.2.32-open-betaOct 25, 2026
No release notes have been provided for this version.
v0.2.31-open-betaOct 22, 2026
Improves basic face centering and dependency checks on Windows and macOS. Presenter windows, captions and narration timing are retained.
View original text
修复 Windows/macOS 粗略固定人脸居中与隔离依赖检查；保持人物框、字幕和口播时钟。
v0.2.30-open-betaOct 22, 2026
Improves portrait narration and captions while preserving the complete content, existing presenter imagery and audio. Updates use the original key; old packages and projects are retained.
View original text
完善竖屏口播与字幕显示，保留完整内容、原有人像与音频。沿用原 Key 更新，保留旧包和工程。
v0.2.29-open-betaOct 20, 2026
Fixed the situation where cropped images were misjudged as covering subtitles, and improved the preview check. Updates reuse the original key; old packages and projects are retained.
View original text
修复裁切图片被误判为遮挡字幕的情况，完善预览检查。沿用原 Key 增量更新，保留旧包和工程。
v0.2.28-open-betaOct 20, 2026
Improved compatibility of authoring tools in different local environments. Updates reuse the original key; old packages and projects are retained.
View original text
改善制作工具在不同本地环境中的兼容性。沿用原 Key 增量更新，保留旧包和工程。
v0.2.27-open-betaOct 19, 2026
Improve subtitles, production parameters and material usage instructions, and improve vertical text layout and production process. Updates reuse the original key; old packages and projects are retained.
View original text
完善字幕、制作参数和素材使用说明，改善竖屏文字排版及制作流程。沿用原 Key 增量更新，保留旧包和工程。
v0.2.26-open-betaOct 19, 2026
Improve the production process, preview and export stability, and improve the assistant usage instructions and exception prompts. Keep the old version, original projects and materials, and continue to use the original Key for incremental updates. The new narrated video supports the use of photos or videos shot by shot with real subtitles, maintaining the original sound, subtitle clock and action sound effects; old projects continue to use the original mode.
View original text
改进制作流程、预览与导出稳定性，完善助手使用说明和异常提示。保留旧版本、原有工程与素材，继续沿用原 Key 进行增量更新。新建口播支持按真实字幕逐镜头使用照片或视频，保持原声、字幕时钟和动作音效；旧工程沿用原模式。
v0.2.25-open-betaOct 19, 2026
Improve the personal template workflow and example guidance, clarify the plan and continue the process after preview review; display the specific reasons when the environment, audio, and export fail to facilitate assistant positioning and recovery. Keep original animations, sound effects, old packages and projects. Instructions for using the unified English vertical screen basic set.
View original text
完善个人包制作入口与示例引导，明确计划和预览审核后的继续流程；环境、音频及导出失败时显示具体原因，便于助手定位和恢复。保留原有动画、音效、旧包与工程。统一英文竖屏基础组的使用说明。
v0.2.24-open-betaOct 18, 2026
Improves scene-to-scene continuity and emphasis, fixes right-edge cropping of basic cards in portrait scenes without a presenter, and improves creation guidance. Existing audio, captions, sound effects, old packages and projects are retained.
View original text
优化逐句画面承接与焦点保持，修复无人物竖屏基础卡片偏右裁切；同步完善新制作引导。保留原声、字幕、原生音效和右上角时间与动态标识，旧包与工程继续保留。
v0.2.23-open-betaOct 17, 2026
Corrected the time verification of automatic subtitle drafts to avoid subtitles and emphasis being advanced; support retaining review records of complete sentence-level subtitles, real word boundary re-segmentation, and audio and non-speech tail segments. The original sound and native sound effects are retained, the original call identifier is used for updates, and the old packages and projects are retained.
View original text
修正自动字幕草稿的时间校验，避免字幕和强调提前；支持保留完整句级字幕、真实词边界重分句及有声非语音尾段的复核记录。保留原声和原生音效，更新沿用原调用标识，旧包与工程保留。
v0.2.22-open-betaOct 17, 2026
Adds English portrait basic expressions B1–B4 and an English portrait demo. Updates reuse the existing key; save the new package separately and keep the old package and projects.
View original text
新增英文竖屏基础表达 B1–B4 及英文竖屏 Demo。更新自动沿用原 Key，新包另存，旧包和已有工程保留。
v0.2.21-open-betaOct 17, 2026
Compatible with abnormal coding output in environmental detection to avoid unexpected interruption of the diagnostic process.
View original text
兼容环境检测中的异常编码输出，避免诊断流程意外中断。
v0.2.20-open-betaOct 17, 2026
Corrected UTF-8 decoding of environment detection text to improve Chinese path and diagnostic output compatibility.
View original text
修正环境检测文本的 UTF-8 解码，改善中文路径与诊断输出兼容性。
v0.2.19-open-betaOct 15, 2026
Optimized the creation workflow, material binding and Chinese and English text layout, repaired Windows relative material paths; added sentence-by-sentence emphasis and independent action markers, improved left and right contrast effects and subtitle display. Retain original animations, sounds, presentations, and supported frames.
View original text
优化制作入口、素材绑定与中英文文字排版，修复 Windows 相对素材路径；新增逐句重点强调和独立行动标记，改善左右对照效果及字幕显示。保留原有动画、声音、演示和已支持画幅。
v0.2.10-open-betaOct 15, 2026
Preserves complete narration and captions: the capacity of one scene does not limit the entire video. The assistant should first split or combine compatible scenes. Existing animation, sound effects, demos, licenses and prices are retained.
View original text
完整口播编排修复：保留完整人声与字幕，单镜头容量不作为整片上限；优先重新分段、组合相容镜头。25个入口及13个包复核通过；原动画、音效、Demo、许可与价格保留。
v0.2.9-open-betaOct 15, 2026
0.2.9 public beta: adds Chinese portrait layout and basic face centering. The Chinese demo is portrait; the English demo remains landscape. The 27 existing animation and sound-effect groups are retained. Full coverage of the original project and English portrait support were not yet complete in this release.
View original text
0.2.9公开测试版：新增中文竖屏及简单人脸居中。中文Demo为竖屏；英文Demo仍为横屏。保留27组现有动作与音效，完整原工程覆盖及英语竖屏仍未完成。
v0.2.7-open-betaOct 15, 2026
0.2.7 public beta, using template address 385712, with Chinese and English guides and demos.
View original text
0.2.7 公开测试版；使用指定站内地址 385712，附中英文指南与 Demo。