Chữ đồng bộ theo âm thanh

Mọi thứ trên trang này đều có bản máy đọc. Đưa agent link JSON, hoặc thêm MCP server gimgs rồi bảo nó tìm motion template.

1. Tải JSON — trong đó có công thức cảnh quay (timing, easing, bảng tham số, lỗi hay gặp) và toàn bộ source TSX.
2. Clone repo video-shotcraft; component import fixture chung từ demos/_fixtures và helper từ assets/lib.
3. Đăng ký component thành <Composition> trong project Remotion rồi render bằng `npx remotion render`, hoặc dựng lại công thức bằng engine của bạn.

curl -s https://gimgs.net/motion/type-rhythm-sync.json

git clone https://github.com/Vincentwei1021/video-shotcraft
# component: demos/typography/type-rhythm-sync/FontWeightPump.tsx, demos/typography/type-rhythm-sync/KaraokeFillSync.tsx
npx remotion render <CompositionId> out.mp4

claude mcp add --transport http gimgs https://gimgs.net/mcp
# then: search_motion_templates / get_motion_template("type-rhythm-sync")

Use the video-shotcraft shot recipe "Chữ đồng bộ theo âm thanh" (Remotion composition, Apache-2.0) published on gimgs.net.
Recipe JSON (prompt = timing/easing/parameter table/pitfalls, files = TSX source): https://gimgs.net/motion/type-rhythm-sync.json
Raw TSX: https://gimgs.net/motion/type-rhythm-sync/template.html
Repository: https://github.com/Vincentwei1021/video-shotcraft  (component: demos/typography/type-rhythm-sync/FontWeightPump.tsx, demos/typography/type-rhythm-sync/KaraokeFillSync.tsx; imports demos/_fixtures and assets/lib)
Spec: 1920x1080, 30 fps, 4.72s

1. Read the JSON and follow its "prompt" section (when to use, intention, parameter table, known pitfalls).
2. Adapt it for: <describe what you want>.
3. Either clone the repo, register the component as a <Composition> in a Remotion project and run npx remotion render <CompositionId> out.mp4 — or re-implement the recipe in your own engine (HTML/WAAPI works: keep the same timing and easing).

JSON · Markdown · llms.txt

// ===== demos/typography/type-rhythm-sync/FontWeightPump.tsx =====
// 字重脉冲（font-weight-pump）——标题笔画随节拍变粗又弹回，像文字跟着低音鼓蹦迪。
// 节拍每 20f 一拍：帧 30/50/70/90/110 命中。命中瞬间 -webkit-text-stroke 0→10px
// 跳满，随后 10f 幂衰减 (1-t/10)^0.8 弹回 0；fontWeight 在命中窗口 400→900 离散跳变
// （env>0.15），配合 stroke 连续衰减读作"连续变粗"。第 3、5 拍（帧 70/110）为重音，
// 额外 scaleX 1→1.08 同衰减（transform 缩放不改排版）。底部 5 个节拍点作节拍参照。
// 结构：0–29f 静止 hold；30–119f 五拍脉冲；120–139f 真静止收尾（20f，110+10=120 无残留）。
import React from 'react';
import { useCurrentFrame } from 'remotion';
import { G } from '../../_fixtures/Fixtures';

const BEATS = [30, 50, 70, 90, 110];
const ACCENTS = new Set([2, 4]); // 第 3、5 拍重音
const DECAY = 10; // 衰减帧数

// 命中后 10f 幂衰减包络：t=0 → 1，t>=10 → 精确 0（保证结尾真静止）
const envAt = (frame: number, beat: number) => {
  const t = frame - beat;
  if (t < 0 || t >= DECAY) return 0;
  return Math.pow(1 - t / DECAY, 0.8);
};

export const FontWeightPump: React.FC = () => {
  const frame = useCurrentFrame();

  // 只可能有一个活跃拍（拍距 20f > 衰减 10f），取最大包络及其拍序号
  let env = 0;
  let activeBeat = -1;
  BEATS.forEach((b, i) => {
    const e = envAt(frame, b);
    if (e > env) {
      env = e;
      activeBeat = i;
    }
  });

  const strokeW = 10 * env; // 笔画粗细连续衰减
  const weight = env > 0.15 ? 900 : 400; // 命中窗口离散跳字重
  const accent = activeBeat >= 0 && ACCENTS.has(activeBeat);
  const scaleX = accent ? 1 + 0.08 * env : 1; // 重音拍变宽一挡

  return (
    <div
      style={{
        width: 1920,
        height: 1080,
        background: G.bg,
        position: 'relative',
        overflow: 'hidden',
        fontFamily: 'Helvetica, Arial, sans-serif',
      }}
    >
      {/* 定宽居中容器，transform 缩放不改排版 */}
      <div
        style={{
          position: 'absolute',
          left: 0,
          top: 0,
          width: 1920,
          height: 1080,
          display: 'flex',
          alignItems: 'center',
          justifyContent: 'center',
        }}
      >
        <div
          style={{
            fontSize: 150,
            fontWeight: weight,
            color: G.ink,
            letterSpacing: 6,
            whiteSpace: 'nowrap',
            WebkitTextStroke: `${strokeW}px ${G.ink}`,
            transform: `scaleX(${scaleX})`,
            transformOrigin: 'center center',
          }}
        >
          PUMP IT UP
        </div>
      </div>

      {/* 底部节拍点：5 个，命中哪拍哪个点闪 ink 并放大 */}
      <div
        style={{
          position: 'absolute',
          left: 0,
          bottom: 130,
          width: 1920,
          display: 'flex',
          justifyContent: 'center',
          gap: 56,
        }}
      >
        {BEATS.map((b, i) => {
          const e = envAt(frame, b);
          const dotOpacity = e > 0.02 ? Math.min(1, 0.3 + e * 1.2) : 0;
          const dotScale = 1 + 0.8 * e;
          return (
            <div
              key={i}
              style={{
                width: 26,
                height: 26,
                borderRadius: 13,
                background: G.line,
                position: 'relative',
                transform: `scale(${dotScale})`,
              }}
            >
              <div
                style={{
                  position: 'absolute',
                  inset: 0,
                  borderRadius: 13,
                  background: G.ink,
                  opacity: dotOpacity,
                }}
              />
            </div>
          );
        })}
      </div>
    </div>
  );
};


// ===== demos/typography/type-rhythm-sync/KaraokeFillSync.tsx =====
// 卡拉OK填色随读（karaoke-fill-sync）——旁白读到哪个词，哪个词就被深色从左到右
// 点亮。两行标语 "SHIP FASTER / BREAK NOTHING"，每个词双层同文本叠放：底层 G.line
// 浅灰字，上层 G.ink 深字用 clip-path: inset(0 X% 0 0) 按词内进度线性填充（逐词独立
// 叠层，clip 百分比即词内进度，无需量测词在整行的像素占比）。词级时间表模拟语速：
// SHIP 20–38、FASTER 42–75（长词慢读）、BREAK 85–103、NOTHING 107–130，词间停顿。
// 正在填的词底下有 8px 深色下划线跟随填充右缘作读指。0–19f hold；130–149f 真静止。
import React from 'react';
import { useCurrentFrame, interpolate } from 'remotion';
import { G } from '../../_fixtures/Fixtures';

type Word = { text: string; start: number; end: number };

const LINES: Word[][] = [
  [
    { text: 'SHIP', start: 20, end: 38 },
    { text: 'FASTER', start: 42, end: 75 },
  ],
  [
    { text: 'BREAK', start: 85, end: 103 },
    { text: 'NOTHING', start: 107, end: 130 },
  ],
];

const KaraokeWord: React.FC<{ word: Word; frame: number }> = ({ word, frame }) => {
  // 词内 linear 填充进度，clamp 保证读完保持
  const p = interpolate(frame, [word.start, word.end], [0, 1], {
    extrapolateLeft: 'clamp',
    extrapolateRight: 'clamp',
  });
  const active = frame >= word.start && frame < word.end; // 正在读这个词
  return (
    <span style={{ position: 'relative', display: 'inline-block' }}>
      {/* 底层：浅灰未读字 */}
      <span style={{ color: G.line }}>{word.text}</span>
      {/* 上层：深字按进度从左到右揭开 */}
      <span
        style={{
          position: 'absolute',
          inset: 0,
          color: G.ink,
          clipPath: `inset(0 ${(1 - p) * 100}% 0 0)`,
        }}
      >
        {word.text}
      </span>
      {/* 读指下划线：只在正在填的词下出现，右缘跟随填充进度 */}
      {active && (
        <span
          style={{
            position: 'absolute',
            left: 0,
            bottom: -14,
            width: `${p * 100}%`,
            height: 8,
            background: G.ink,
          }}
        />
      )}
    </span>
  );
};

export const KaraokeFillSync: React.FC = () => {
  const frame = useCurrentFrame();
  return (
    <div
      style={{
        width: 1920,
        height: 1080,
        background: G.bg,
        overflow: 'hidden',
        display: 'flex',
        flexDirection: 'column',
        justifyContent: 'center',
        paddingLeft: 240,
        boxSizing: 'border-box',
        fontFamily: 'Helvetica, Arial, sans-serif',
        fontSize: 130,
        fontWeight: 800,
        letterSpacing: 2,
        lineHeight: 1.45,
      }}
    >
      {LINES.map((words, li) => (
        <div key={li} style={{ display: 'flex', gap: 48 }}>
          {words.map((w) => (
            <KaraokeWord key={w.text} word={w} frame={frame} />
          ))}
        </div>
      ))}
    </div>
  );
};

1. Read the recipe
   Prompt = when to use, intention, parameter table with typical values and how each one feels, known pitfalls. Treat values as calibrated starting points.
2. Get the code
   tool: git
   git clone https://github.com/Vincentwei1021/video-shotcraft && cd video-shotcraft && npm install
3. Register + render
   tool: remotion
   npx remotion render <CompositionId> out.mp4 # component: demos/typography/type-rhythm-sync/FontWeightPump.tsx, demos/typography/type-rhythm-sync/KaraokeFillSync.tsx
4. Or re-implement
   Port the timing/easing from the recipe to HTML + WAAPI (gimgs motion contract) or any engine; keep the same segment windows and easing.