tts-voice-creator-clone-and.../static/js/audiobook-segmentation.js

64 lines
2.8 KiB
JavaScript

// ── Audiobook quote segmentation ────────────────────────────────────────────
// Quote-boundary segmentation and conservative turn-taking are kept separate
// from the live LLM casting workflow. Their shared quote/heuristic helpers are
// resolved when a passage is processed, preserving the legacy global API.
function _abImplausibleDialoguePiece(text) {
const t = String(text || '').trim();
if (!t) return true;
const words = t.split(/\s+/).filter(Boolean);
return words.length === 1 && !/[.!?…]["'’”»]?$/.test(t);
}
function audiobookSplitByQuotes(text) {
const spans = [];
for (let i = 0; i < text.length; i++) {
const open = text[i], close = AB_QUOTE_PAIRS[open];
if (!close) continue;
const minStart = spans.length ? spans[spans.length - 1].end : 0;
const orphan = _abOrphanClosingQuoteSpan(text, i, minStart);
if (orphan) { spans.push(orphan); i = orphan.end - 1; continue; }
if ((open === '«' || open === '»') && i > 0 && !/[\s([{—–-]/.test(text[i - 1])) continue;
const raw = text.slice(i + 1);
const closeIdx = raw.indexOf(close);
const endInRaw = _abUnclosedQuoteEnd(raw, closeIdx);
const quote = raw.slice(0, endInRaw).trim();
if (!quote) continue;
const consumedClose = closeIdx >= 0 && endInRaw === closeIdx;
spans.push({ start: i, end: i + 1 + endInRaw + (consumedClose ? 1 : 0), quote });
i = spans[spans.length - 1].end - 1;
}
if (!spans.length) return [{ speaker: 'Narrator', type: 'narration', text, emotion: '' }];
const out = []; let last = 0;
for (let k = 0; k < spans.length; k++) {
const sp = spans[k], pre = text.slice(last, sp.start);
if (pre.trim()) out.push({ speaker: 'Narrator', type: 'narration', text: pre.trim(), emotion: '' });
if (sp.quote) {
const after = text.slice(sp.end, k + 1 < spans.length ? spans[k + 1].start : text.length);
out.push({ speaker: audiobookGuessSpeaker(after, pre) || 'Unknown', type: 'dialogue', text: sp.quote, emotion: '' });
}
last = sp.end;
}
const tail = text.slice(last).trimEnd();
if (tail.trim()) out.push({ speaker: 'Narrator', type: 'narration', text: tail.trim(), emotion: '' });
return audiobookTurnTaking(out);
}
function audiobookTurnTaking(segs) {
let a = null, b = null;
for (const s of segs) {
if (s.type !== 'dialogue') continue;
if (s.speaker && !/^Unknown|Unbekannt/i.test(s.speaker)) {
if (s.speaker !== b) { a = b; b = s.speaker; }
} else if (a && b && a !== b) {
s.speaker = a;
const t = a; a = b; b = t;
}
}
return segs;
}
window.TTVC = window.TTVC || {};
window.TTVC.audiobook = window.TTVC.audiobook || {};
window.TTVC.audiobook.segmentation = { splitByQuotes: audiobookSplitByQuotes, turnTaking: audiobookTurnTaking };