Visual UI for Explainroo. Explainer videos and product demos made by your AI agent. Free and open source: a local voice (Kokoro), word timing (Whisper) and a canvas renderer turn a script into a narrated MP4.
import React, { useState, useRef, useEffect } from 'react';
import {
Volume2,
Sparkles,
Clock,
Check,
RefreshCw,
Upload,
Trash2,
Mic,
} from 'lucide-react';
import { Scene, VisualElement } from '../../types/explainroo';
import { audioEngine } from '../../utils/audioEngine';
interface ScriptEditorProps {
scene: Scene;
selectedElementId: string | null;
onUpdateNarration: (text: string) => void;
onUpdateElementTriggerWord: (elementId: string, word: string) => void;
onUpdateCustomAudio?: (url?: string, filename?: string) => void;
voiceRate: number;
voicePitch: number;
}
export const ScriptEditor: React.FC<ScriptEditorProps> = ({
scene,
selectedElementId,
onUpdateNarration,
onUpdateElementTriggerWord,
onUpdateCustomAudio,
voiceRate,
voicePitch,
}) => {
const [isPolishing, setIsPolishing] = useState(false);
const [isPlayingVoice, setIsPlayingVoice] = useState(false);
const [liveSpokenWordIndex, setLiveSpokenWordIndex] = useState<number>(-1);
// Subscribe to real-time word boundary updates from AudioEngine
useEffect(() => {
const unsub = audioEngine.addWordListener((_word, _charIdx, wordIdx) => {
setLiveSpokenWordIndex(wordIdx);
if (wordIdx < 0) {
setIsPlayingVoice(false);
}
});
return () => unsub();
}, []);
const words = scene.narration
.split(/\s+/)
.filter((w) => w.length > 0)
.map((rawWord) => rawWord.replace(/[.,/#!$%^&*;:{}=\-_`~()?"']/g, ''));
const wordCount = words.length;
// Estimated reading time at ~150 words per minute (2.5 words per second)
const estimatedSeconds = Math.max(3, Math.round(wordCount / 2.5));
// Find currently selected element
const selectedElement = scene.elements.find((el) => el.id === selectedElementId);
// Map each word to elements that trigger on it
const wordTriggerMap: Record<string, VisualElement[]> = {};
scene.elements.forEach((el) => {
if (el.triggerWord) {
const cleanTrigger = el.triggerWord.toLowerCase().replace(/[^a-z0-9]/g, '');
if (!wordTriggerMap[cleanTrigger]) {
wordTriggerMap[cleanTrigger] = [];
}
wordTriggerMap[cleanTrigger].push(el);
}
});
const handlePolishWithAi = async () => {
if (!scene.narration) return;
try {
setIsPolishing(true);
const res = await fetch('/api/explainroo/improve-script', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ text: scene.narration, goal: 'punchy and conversational' }),
});
if (res.ok) {
const data = await res.json();
if (data.improvedText) {
onUpdateNarration(data.improvedText);
audioEngine.playDrawChime();
}
}
} catch (err) {
console.error('Failed to polish script:', err);
} finally {
setIsPolishing(false);
}
};
const fileInputRef = useRef<HTMLInputElement | null>(null);
const handleAudioUpload = (e: React.ChangeEvent<HTMLInputElement>) => {
const file = e.target.files?.[0];
if (!file) return;
const url = URL.createObjectURL(file);
if (onUpdateCustomAudio) {
onUpdateCustomAudio(url, file.name);
}
audioEngine.playPop();
};
const handleRemoveCustomAudio = () => {
if (onUpdateCustomAudio) {
onUpdateCustomAudio(undefined, undefined);
}
audioEngine.stopCustomAudio();
setIsPlayingVoice(false);
};
const handleTestVoice = () => {
if (isPlayingVoice) {
audioEngine.stopSpeaking();
setIsPlayingVoice(false);
setLiveSpokenWordIndex(-1);
} else {
setIsPlayingVoice(true);
if (scene.customAudioUrl) {
audioEngine.playCustomAudio(scene.customAudioUrl, () => {
setIsPlayingVoice(false);
setLiveSpokenWordIndex(-1);
});
} else {
audioEngine.speakNarration(
scene.narration,
(_word, _charIdx, wordIdx) => {
setLiveSpokenWordIndex(wordIdx);
},
() => {
setIsPlayingVoice(false);
setLiveSpokenWordIndex(-1);
},
voiceRate,
voicePitch
);
}
}
};
const handleWordChipClick = (wordText: string) => {
if (!selectedElement) return;
const clean = wordText.toLowerCase();
onUpdateElementTriggerWord(selectedElement.id, clean);
audioEngine.playPop();
};
return (
<div className="bg-slate-900 border border-slate-800 rounded-2xl p-4 flex flex-col gap-3">
{/* Header */}
<div className="flex items-center justify-between pb-2 border-b border-slate-800">
<div>
<h3 className="text-xs font-bold uppercase tracking-wider text-slate-300">
Voiceover Script & Word Sync
</h3>
<p className="text-[11px] text-slate-500">
Explainroo synchronizes your drawings with spoken words
</p>
</div>
<div className="flex items-center gap-2">
{/* Audio Test Voice */}
<button
onClick={handleTestVoice}
className={`flex items-center gap-1 px-2.5 py-1 text-xs font-medium rounded-lg border transition-colors cursor-pointer ${
isPlayingVoice
? 'bg-amber-500/20 text-amber-300 border-amber-500/40'
: 'bg-slate-800 text-slate-300 border-slate-700 hover:text-white'
}`}
title="Preview spoken narration"
>
<Volume2 className="w-3.5 h-3.5 text-amber-400" />
<span>{isPlayingVoice ? 'Speaking...' : 'Listen'}</span>
</button>
{/* AI Polish Script */}
<button
onClick={handlePolishWithAi}
disabled={isPolishing || !scene.narration}
className="flex items-center gap-1 px-2.5 py-1 bg-amber-500/15 hover:bg-amber-500/25 border border-amber-500/30 text-amber-300 text-xs font-semibold rounded-lg transition-colors cursor-pointer disabled:opacity-50"
title="Make script punchier using AI"
>
{isPolishing ? (
<RefreshCw className="w-3.5 h-3.5 animate-spin" />
) : (
<Sparkles className="w-3.5 h-3.5" />
)}
<span>AI Polish</span>
</button>
</div>
</div>
{/* Narration Text Area */}
<div className="relative">
<textarea
rows={3}
value={scene.narration}
onChange={(e) => onUpdateNarration(e.target.value)}
placeholder="Write the narration voiceover for this scene..."
className="w-full bg-slate-950 border border-slate-800 focus:border-cyan-400/80 rounded-xl p-3 text-xs sm:text-sm text-slate-200 outline-none resize-none transition-colors"
/>
<div className="flex items-center justify-between text-[11px] text-slate-500 px-1 mt-1">
<span className="flex items-center gap-1">
<Clock className="w-3 h-3 text-slate-400" />
<span>~{estimatedSeconds}s spoken ({wordCount} words)</span>
</span>
<span className="italic text-slate-400">
{selectedElement
? `Click a word below to anchor "${selectedElement.label || selectedElement.type}"`
: 'Select an element to sync it to a word'}
</span>
</div>
</div>
{/* Custom Audio Upload Row */}
<div className="flex items-center justify-between p-2 rounded-xl bg-slate-950/70 border border-slate-800/80">
<div className="flex items-center gap-2 overflow-hidden">
<input
ref={fileInputRef}
type="file"
accept="audio/*"
onChange={handleAudioUpload}
className="hidden"
/>
<Mic className="w-3.5 h-3.5 text-slate-400 shrink-0" />
{scene.customAudioUrl ? (
<div className="flex items-center gap-1.5 truncate">
<span className="w-2 h-2 rounded-full bg-emerald-400 shrink-0 animate-pulse" />
<span className="text-xs text-emerald-300 font-mono truncate max-w-[180px]">
{scene.customAudioName || 'Custom Audio'}
</span>
</div>
) : (
<span className="text-xs text-slate-400">
Web Speech TTS active · Or upload voiceover file
</span>
)}
</div>
<div className="flex items-center gap-1.5 shrink-0">
{scene.customAudioUrl ? (
<button
onClick={handleRemoveCustomAudio}
className="p-1 rounded-lg text-slate-400 hover:text-rose-400 hover:bg-slate-800 transition-colors cursor-pointer"
title="Remove custom audio and use Web Speech TTS"
>
<Trash2 className="w-3.5 h-3.5" />
</button>
) : (
<button
onClick={() => fileInputRef.current?.click()}
className="inline-flex items-center gap-1 px-2 py-1 rounded-lg bg-slate-800 hover:bg-slate-700 text-slate-200 text-xs font-medium border border-slate-700 transition-colors cursor-pointer"
title="Upload MP3 or WAV voiceover track"
>
<Upload className="w-3 h-3 text-slate-400" />
<span>Upload MP3/WAV</span>
</button>
)}
</div>
</div>
{/* Interactive Word Sync Chips */}
{words.length > 0 && (
<div className="pt-2">
<span className="text-[10px] font-semibold tracking-wider uppercase text-slate-400 block mb-2">
Click Word to Anchor Selected Element:
</span>
<div className="flex flex-wrap gap-1.5 max-h-36 overflow-y-auto pr-1">
{words.map((w, idx) => {
const clean = w.toLowerCase();
const isAttachedToSelected =
selectedElement &&
selectedElement.triggerWord?.toLowerCase().replace(/[^a-z0-9]/g, '') === clean;
const attachedElements = wordTriggerMap[clean] || [];
const isSpeakingNow = liveSpokenWordIndex === idx;
return (
<button
key={`${w}-${idx}`}
onClick={() => handleWordChipClick(w)}
className={`inline-flex items-center gap-1 px-2 py-1 rounded-lg text-xs font-medium transition-all cursor-pointer border ${
isSpeakingNow
? 'bg-amber-400 text-slate-950 font-black border-amber-300 scale-110 shadow-[0_0_18px_rgba(251,191,36,0.95)] ring-2 ring-amber-300 z-10 animate-pulse'
: isAttachedToSelected
? 'bg-amber-400 text-slate-950 font-bold border-amber-300 shadow-[0_0_12px_rgba(251,191,36,0.4)] scale-105'
: attachedElements.length > 0
? 'bg-cyan-950/60 border-cyan-500/40 text-cyan-300 hover:border-cyan-400'
: 'bg-slate-950/60 border-slate-800/80 text-slate-300 hover:border-slate-700 hover:bg-slate-800/60'
}`}
title={
isSpeakingNow
? `Currently speaking: "${w}"`
: selectedElement
? `Click to trigger "${selectedElement.label || selectedElement.type}" on "${w}"`
: `${attachedElements.length} element(s) trigger here`
}
>
<span>{w}</span>
{attachedElements.length > 0 && !isAttachedToSelected && !isSpeakingNow && (
<span className="w-1.5 h-1.5 rounded-full bg-cyan-400" />
)}
{isAttachedToSelected && !isSpeakingNow && (
<Check className="w-3 h-3 text-slate-950 stroke-[3]" />
)}
{isSpeakingNow && (
<span className="w-1.5 h-1.5 rounded-full bg-slate-950 animate-ping" />
)}
</button>
);
})}
</div>
</div>
)}
</div>
);
};