125 lines
4.5 KiB
JavaScript
125 lines
4.5 KiB
JavaScript
// ---------- Сравнение текстов (Левенштейн) и doCompare ----------
|
||
|
||
function normalize(text) {
|
||
return text.toLowerCase().replace(/[.,!?;:«»"'’]/g, '').trim();
|
||
}
|
||
|
||
function levenshtein(a, b) {
|
||
const m = a.length, n = b.length;
|
||
const dp = Array.from({length: m+1}, (_,i) => [i]);
|
||
for (let j=0; j<=n; j++) dp[0][j] = j;
|
||
for (let i=1; i<=m; i++)
|
||
for (let j=1; j<=n; j++)
|
||
dp[i][j] = a[i-1] === b[j-1] ? dp[i-1][j-1] : 1 + Math.min(dp[i-1][j], dp[i][j-1], dp[i-1][j-1]);
|
||
return dp[m][n];
|
||
}
|
||
|
||
function wordSimilarity(a, b) {
|
||
const maxLen = Math.max(a.length, b.length);
|
||
if (maxLen === 0) return 1;
|
||
return 1 - levenshtein(a, b) / maxLen;
|
||
}
|
||
|
||
function compareTexts(original, transcribed) {
|
||
const a = normalize(original).split(/\s+/);
|
||
const b = normalize(transcribed).split(/\s+/);
|
||
let totalScore = 0;
|
||
const result = a.map(word => {
|
||
let best = 0, bestMatch = '';
|
||
for (const t of b) {
|
||
const sim = wordSimilarity(word, t);
|
||
if (sim > best) { best = sim; bestMatch = t; }
|
||
}
|
||
totalScore += best;
|
||
const quality = best >= 0.8 ? 'ok' : best >= 0.5 ? 'near' : 'bad';
|
||
return { word, quality, bestMatch, score: Math.round(best*100) };
|
||
});
|
||
const score = Math.round((totalScore / a.length) * 100);
|
||
return { score, words: result };
|
||
}
|
||
|
||
function renderDiff(words) {
|
||
return words.map(({ word, quality, bestMatch, score }) => {
|
||
const color = quality === 'ok' ? 'green' : quality === 'near' ? '#e6a800' : 'red';
|
||
const tip = quality !== 'ok' ? ` → ${bestMatch} (${score}%)` : '';
|
||
return `<span style="color:${color}; font-weight:bold">${word}</span><span style="color:#888; font-size:0.85em">${tip}</span>`;
|
||
}).join(' ');
|
||
}
|
||
|
||
async function doCompare(blob, originalText, duration) {
|
||
if (isAnalyzing) return '';
|
||
isAnalyzing = true;
|
||
const startTime = Date.now();
|
||
const timerId = setInterval(() => {
|
||
const elapsed = ((Date.now() - startTime) / 1000).toFixed(1);
|
||
setStatus('Анализ... ' + elapsed + 'с');
|
||
}, 150);
|
||
let whisperResult;
|
||
try {
|
||
whisperResult = await transcribe(blob);
|
||
} catch(e) {
|
||
clearInterval(timerId);
|
||
isAnalyzing = false;
|
||
setStatus('Ошибка анализа.');
|
||
return '';
|
||
}
|
||
clearInterval(timerId);
|
||
|
||
let recHTML = '', diffHTML = '';
|
||
|
||
if (whisperResult.text) {
|
||
const _transcribed = whisperResult.text;
|
||
const { score, words } = compareTexts(originalText, whisperResult.text);
|
||
const recEmoji = score >= 90 ? '🟢' : score >= 70 ? '🟡' : '🔴';
|
||
recHTML = recEmoji + ' Распознание: <b>' + score + '%</b> | 🗣 <i>' + whisperResult.text + '</i>';
|
||
diffHTML =
|
||
'<div style="margin:6px 0;font-size:0.8em;color:#999">' +
|
||
'📋 Текст: зелёный ≥80%, жёлтый ≥50%, красный <50%' +
|
||
'</div>' +
|
||
renderDiff(words);
|
||
} else {
|
||
recHTML = '⚠️ Не распознано';
|
||
diffHTML = '';
|
||
}
|
||
|
||
document.getElementById('score').innerHTML = recHTML;
|
||
|
||
document.getElementById('diff').innerHTML = diffHTML;
|
||
|
||
// Слоги — в отдельный div, не в diff (diff перезаписывается)
|
||
_currentBlob = blob;
|
||
_currentSylBuf = null;
|
||
_currentSylMap = whisperResult.words && whisperResult.words.length > 0
|
||
? mapSyllablesToTimestamps(originalText, whisperResult.words)
|
||
: null;
|
||
console.log('[SYL] map:', _currentSylMap);
|
||
const sylDiv = document.getElementById('syllables');
|
||
sylDiv.innerHTML = '';
|
||
const sy = syllabifyIT(originalText);
|
||
if (sy && sy.includes('-')) {
|
||
const syls = sy.split('-');
|
||
const div = document.createElement('div');
|
||
div.style.cssText = 'margin-top:8px;font-size:1.2em;letter-spacing:2px;user-select:none';
|
||
syls.forEach((s, i) => {
|
||
if (i > 0) {
|
||
const sep = document.createElement('span');
|
||
sep.style.cssText = 'color:#ccc;margin:0 1px';
|
||
sep.textContent = '-';
|
||
div.appendChild(sep);
|
||
}
|
||
const chip = document.createElement('span');
|
||
chip.textContent = s;
|
||
chip.style.cssText = 'cursor:pointer;padding:2px 6px;border-radius:4px;transition:background 0.1s';
|
||
chip.onmouseover = () => { chip.style.background = '#d0d0ff'; };
|
||
chip.onmouseout = () => { chip.style.background = ''; };
|
||
chip.onclick = () => { chip.style.background = '#a0a0ff'; setTimeout(() => { chip.style.background = '#d0d0ff'; }, 200); playSyllable(i, syls.length); };
|
||
div.appendChild(chip);
|
||
});
|
||
sylDiv.appendChild(div);
|
||
}
|
||
|
||
setStatus('Готово.');
|
||
isAnalyzing = false;
|
||
return whisperResult.text || '';
|
||
}
|