-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathlinglu-en.html
More file actions
656 lines (630 loc) · 29.2 KB
/
Copy pathlinglu-en.html
File metadata and controls
656 lines (630 loc) · 29.2 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
<!doctype html>
<html lang="en">
<head>
<meta charset="UTF-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>LingLu · Live Topic Tree - Boundless Flow</title>
<link rel="stylesheet" href="style.css" />
</head>
<body>
<div class="sidebar">
<a href="welcome-en.html" class="sidebar-logo">
<svg
width="24"
height="24"
viewBox="0 0 24 24"
fill="none"
stroke="currentColor"
stroke-width="2"
stroke-linecap="round"
stroke-linejoin="round"
style="color: var(--primary-color)"
>
<path
d="M12 2a3 3 0 0 0-3 3v7a3 3 0 0 0 6 0V5a3 3 0 0 0-3-3Z"
></path>
<path d="M19 10v2a7 7 0 0 1-14 0v-2"></path>
<line x1="12" y1="19" x2="12" y2="22"></line>
</svg>
Boundless Flow
</a>
<div class="sidebar-group">Getting Started</div>
<ul>
<li><a href="welcome-en.html">What is Boundless Flow?</a></li>
<li><a href="onboarding-en.html">Onboarding Wizard</a></li>
</ul>
<div class="sidebar-group">Core Features</div>
<ul>
<li><a href="stt-en.html">Real-time STT & Models</a></li>
<li><a href="translation-en.html">Real-time Translation</a></li>
<li>
<a href="proofreading-summary-en.html"
>AI Proofreading & Summary</a
>
</li>
<li>
<a href="tts-voice-cloning-en.html"
>TTS & Voice Cloning</a
>
</li>
<li><a href="sts-en.html">STS Speech Workbench</a></li>
<li>
<a href="linglu-en.html" class="active"
>LingLu · Live Topic Tree</a
>
</li>
</ul>
<div class="sidebar-group">Appendix</div>
<ul>
<li><a href="appendix-en.html">Beginner Guide</a></li>
</ul>
<div
style="
margin-top: auto;
padding-top: 1rem;
border-top: 1px solid var(--border-color);
"
>
<a
href="linglu.html"
style="display: flex; align-items: center; gap: 0.5rem"
>
<svg
width="16"
height="16"
viewBox="0 0 24 24"
fill="none"
stroke="currentColor"
stroke-width="2"
>
<circle cx="12" cy="12" r="10"></circle>
<line x1="2" y1="12" x2="22" y2="12"></line>
<path
d="M12 2a15.3 15.3 0 0 1 4 10 15.3 15.3 0 0 1-4 10 15.3 15.3 0 0 1-4-10 15.3 15.3 0 0 1 4-10z"
></path>
</svg>
中文版
</a>
</div>
</div>
<div class="main-content">
<div class="content-wrapper">
<div class="page-kicker">
<span class="kicker-dot"></span>
<span>LINGLU · LIVE TOPIC TREE</span>
<span class="version-badge">v0.4</span>
</div>
<h1>LingLu · Live Topic Tree Minutes</h1>
<p class="hero-subtitle">
A live topic tree grows during recording; HTML reports
export when you stop. On-device first (Ollama recommended),
9 LLM providers, 5 report themes.
</p>
<ul class="feature-pill-list">
<li class="accent">
Live mode: topic tree + "now speaking"
</li>
<li class="violet">
Wrap-up: summary / decisions / actions
</li>
<li class="teal">Delivery: 5 HTML report themes</li>
<li class="warm">9 LLM providers, Ollama recommended</li>
</ul>
<h2>Three states, one data model</h2>
<p>
LingLu splits a recording session into three coordinated
states sharing a single <code>LingLuDoc</code>:
</p>
<div class="card-grid">
<div class="card">
<div class="card-icon">🌿</div>
<h3>Live · while recording</h3>
<p>
The "LingLu" output panel grows a live topic tree in
the dock area. Closed topics never get rewritten;
the current open topic is highlighted with a
vermilion left border; the "now speaking" zone shows
the active partial.
</p>
</div>
<div class="card">
<div class="card-icon">📜</div>
<h3>Wrap-up · after stop</h3>
<p>
Auto-consolidates summary / key decisions / action
items. The scenario field decides whether
business-style fields are emitted (personal /
gathering scenarios never force-extract action
items).
</p>
</div>
<div class="card">
<div class="card-icon">📤</div>
<h3>Delivery · pick a theme</h3>
<p>
Toolbar theme dropdown →
<code>👁 Preview</code> (iframe sandbox) or
<code>⬇ Export</code> (lands in
<code>session/<id>/reports/</code>). Five
themes: terminal-grade / keynote-stage / editorial /
minimal-print / sumi-ink.
</p>
</div>
</div>
<h2>9 supported LLM providers</h2>
<p>
Boundless Flow is an <strong>on-device product</strong>, so
Ollama (local) is the default and recommended path — no API
key, zero cost per call, data never leaves your machine. The
other 8 are cloud options and
<strong
>all support custom third-party OpenAI-compatible proxy
URLs</strong
>.
</p>
<div class="card-grid">
<div class="card">
<h3>🏠 Ollama · local (recommended)</h3>
<p>
Default base URL
<code>http://localhost:11434</code>; no API key
required. Also accepts remote Ollama like
<code>http://192.168.x.x:11434</code>.
</p>
</div>
<div class="card">
<h3>☁️ OpenAI-compatible</h3>
<p>
DeepSeek / Kimi (Moonshot) / GLM (Zhipu) /
Volcengine (Doubao) / MiniMax / OpenAI itself — 6
providers share the same OpenAI-compatible codec.
</p>
</div>
<div class="card">
<h3>☁️ Anthropic Claude</h3>
<p>
Messages API (<code>/v1/messages</code>) with its
own codec; <code>x-api-key</code> +
<code>anthropic-version: 2023-06-01</code>;
<code>max_tokens</code> required.
</p>
</div>
<div class="card">
<h3>☁️ Google Gemini</h3>
<p>
v1beta <code>generateContent</code> /
<code>streamGenerateContent?alt=sse</code>;
<code>x-goog-api-key</code> auth;
<code>systemInstruction</code> +
<code>contents</code> body shape.
</p>
</div>
</div>
<h2>Eight scenarios · auto detection</h2>
<p>
LingLu swaps prompt strategy by scenario, producing visibly
different output styles:
</p>
<div class="config-table">
<table>
<thead>
<tr>
<th>Code</th>
<th>Scenario</th>
<th>Default theme</th>
<th>Tone</th>
</tr>
</thead>
<tbody>
<tr>
<td>S0</td>
<td><strong>generic</strong> · fallback</td>
<td>minimal-print</td>
<td>
Neutral note-taking; never assumes a
scenario. auto falls back to this when
confidence < 0.5
</td>
</tr>
<tr>
<td>S1</td>
<td>
<strong>personal</strong> · solo recording ★
</td>
<td>sumi-ink</td>
<td>
Classical Chinese literati voice ("thinking
of…", "feeling…"); classical chapter names
(录略 / 念想 / 断语)
</td>
</tr>
<tr>
<td>S2</td>
<td><strong>private_chat</strong> · 1-on-1</td>
<td>minimal-print</td>
<td>
Two-sided positions + commitment phrases ("I
will…", "I promise…")
</td>
</tr>
<tr>
<td>S3</td>
<td>
<strong>gathering</strong> · informal
gathering ★
</td>
<td>sumi-ink</td>
<td>
Chinese literati gathering tone; small talk
OK; no action items
</td>
</tr>
<tr>
<td>S4</td>
<td>
<strong>closed_meeting</strong> ·
closed-door
</td>
<td>terminal-grade</td>
<td>
Verb-noun topics; prioritize decision verbs
(decide / approve / agree)
</td>
</tr>
<tr>
<td>S5</td>
<td>
<strong>public_meeting</strong> · public
</td>
<td>keynote-stage</td>
<td>
Agenda-style topics; emphasize public
commitments
</td>
</tr>
<tr>
<td>S6</td>
<td><strong>business</strong> · KPI/sales</td>
<td>terminal-grade</td>
<td>
KPIs / projects / clients first; preserve
numbers / amounts / dates
</td>
</tr>
<tr>
<td>S7</td>
<td>
<strong>academic</strong> · research talk
</td>
<td>editorial</td>
<td>
Claims / evidence / counter-arguments /
citations
</td>
</tr>
</tbody>
</table>
</div>
<p>
★ personal & gathering use the
<strong>sumi-ink</strong> theme (Song-dynasty palette:
paper-white + vermilion + Songti/Kaiti typefaces). Content
stays half-classical half-vernacular; business jargon is
forbidden.
</p>
<div class="callout">
<div class="callout-icon">💡</div>
<div>
<strong>auto detection rule:</strong> with scenario =
<code>auto</code>, after ~90s of transcript, the system
runs one cheap LLM call for classification.
<strong
>If confidence < 0.5, the result is forcibly
downgraded to <code>generic</code></strong
>
— this keeps casual / private recordings from being
misclassified into "business commitments".
</div>
</div>
<h2>Five report themes</h2>
<div class="card-grid">
<div class="card">
<h3>🖥 terminal-grade</h3>
<p>
Serious execution; deep #0b0d10 ground + warm gold
accent. KPI cards + ASCII topic tree + OPEN/CLOSED
status.
<strong
>Default for S4 closed / S6 business.</strong
>
</p>
</div>
<div class="card">
<h3>🎤 keynote-stage</h3>
<p>
Stage style; deep-blue gradient + gold + teal;
centered hero + three-pillar conclusions.
<strong>Default for S5 public.</strong>
</p>
</div>
<div class="card">
<h3>📰 editorial</h3>
<p>
Magazine feel; paper white + vermilion; serif
headlines + two-column abstract + roman-numeral
topics + large chapter numbers.
<strong>Default for S7 academic.</strong>
</p>
</div>
<div class="card">
<h3>📄 minimal-print</h3>
<p>
Quiet, A4-friendly; pure white + gray text + no
accent color; <code>@page</code> print styles.
<strong
>Default for S0 generic / S2
private_chat.</strong
>
</p>
</div>
<div class="card">
<h3>🖋 sumi-ink · Chinese aesthetic</h3>
<p>
Song-dynasty palette: paper #f4ede0 + ink #1b1a17 +
vermilion #8a3b2e; Kaiti chapter names (题记 / 录略
/ 念想 / 断语 / 待办 / 跋).
<strong
>Default for S1 personal / S3 gathering.</strong
>
</p>
</div>
</div>
<p class="hero-subtitle" style="font-size: 0.95rem">
All themes inject <code>Content-Security-Policy</code> meta
and forbid <code><script></code> / external CSS /
external images / iframes; every LLM/transcript field is
HTML-escaped; self-contained < 200 KB; mobile-safe at
< 640px.
</p>
<h2>Quick start</h2>
<div class="steps">
<div class="step">
<div class="step-number">1</div>
<div class="step-content">
<h3>Onboarding (first install)</h3>
<p>
Go to <strong>Step 4 · API Providers</strong>,
scroll down to "<strong
>LingLu API · Live topic tree +
reports</strong
>". Defaults are Ollama +
<code>http://localhost:11434</code>; just fill
in the model name (e.g.
<code>qwen3.6:35b</code>).
</p>
</div>
</div>
<div class="step">
<div class="step-number">2</div>
<div class="step-content">
<h3>Main panel · Advanced (anytime)</h3>
<p>
Settings → Advanced →
<strong
>LingLu API · Live topic tree +
reports</strong
>
(open by default). Click
<strong>"Fill defaults"</strong> to auto-fill
base URL by provider; click
<strong>"Test connection"</strong> to verify in
one shot.
</p>
</div>
</div>
<div class="step">
<div class="step-number">3</div>
<div class="step-content">
<h3>Start recording</h3>
<p>
Press <code>RightAlt</code> or click
<code>Start Recording</code>; open
<strong>AI Meeting View</strong> and watch the
dock area "Live topic tree" grow. Click
<strong>↗ Floating Window</strong> to detach to
a side screen.
</p>
</div>
</div>
<div class="step">
<div class="step-number">4</div>
<div class="step-content">
<h3>Export a themed report</h3>
<p>
Pick a theme from the dock toolbar (smart
default by scenario) →
<code>👁 Preview</code> or
<code>⬇ Export</code>. Reports land in
<code
>session/<id>/reports/<theme>.html</code
>
and persist across restarts.
</p>
</div>
</div>
</div>
<h2>Configuration reference</h2>
<div class="config-table">
<table>
<thead>
<tr>
<th>Field</th>
<th>Meaning</th>
<th>Example</th>
</tr>
</thead>
<tbody>
<tr>
<td>Provider</td>
<td>9 options; Ollama listed first</td>
<td><code>ollama</code></td>
</tr>
<tr>
<td>Base URL</td>
<td>
Any OpenAI-compatible proxy works (e.g.
oneapi.example.com); Ollama defaults to
localhost
</td>
<td><code>http://localhost:11434</code></td>
</tr>
<tr>
<td>API Key</td>
<td>
Leave empty for Ollama; required otherwise
</td>
<td><code>sk-xxxxxxxx</code></td>
</tr>
<tr>
<td>Model</td>
<td>Model id with tag</td>
<td>
<code>qwen3.6:35b</code> /
<code>deepseek-chat</code> /
<code>claude-3-5-haiku-20241022</code>
</td>
</tr>
<tr>
<td>Scenario</td>
<td>
auto + 8 scenarios; auto with low confidence
falls back to generic
</td>
<td><code>auto</code></td>
</tr>
<tr>
<td>Report theme</td>
<td>5 themes; smart default by scenario</td>
<td><code>sumi-ink</code></td>
</tr>
<tr>
<td>Temperature</td>
<td>
0.0–1.0; 0.1–0.3 recommended for minutes
</td>
<td><code>0.2</code></td>
</tr>
<tr>
<td>Max tokens</td>
<td>
≥ 2048 recommended for reasoning models
(chain-of-thought eats budget)
</td>
<td><code>4096</code></td>
</tr>
<tr>
<td>Interval (seconds)</td>
<td>
Lower bound 10, default 60; shorter = more
real-time but more calls
</td>
<td><code>60</code></td>
</tr>
</tbody>
</table>
</div>
<h2>Session directory layout</h2>
<p>
After stop, artifacts land under
<code>sessions/<YYYYMMDD_HHMMSS_slug>/</code>:
</p>
<div class="config-table">
<table>
<thead>
<tr>
<th>File</th>
<th>Purpose</th>
</tr>
</thead>
<tbody>
<tr>
<td><code>audio.wav</code></td>
<td>Source recording (existing)</td>
</tr>
<tr>
<td><code>enriched_transcript.json</code></td>
<td>Speaker-tagged transcript (existing)</td>
</tr>
<tr>
<td><code>minutes.json</code></td>
<td>
Rule-based minutes (existing) + new LingLu
fields: topic_tree_ref / scenario / theme_id
/ llm_provider / llm_model / linglu_schema
</td>
</tr>
<tr>
<td><code>minutes.html</code></td>
<td>
Minutes HTML —
<strong
>v0.4 now renders with the
scenario-appropriate theme</strong
>
(no more blue-teal gradient)
</td>
</tr>
<tr>
<td><code>linglu.json</code></td>
<td>
<strong>NEW</strong>: full LingLuDoc with
topics / liveTail / runtime /
providerSnapshot
</td>
</tr>
<tr>
<td><code>reports/<theme>.html</code></td>
<td>
<strong>NEW</strong>: user-exported themed
reports
</td>
</tr>
</tbody>
</table>
</div>
<h2>Security & privacy</h2>
<ul>
<li>
<strong>No raw audio sent</strong>: only transcript +
speaker labels + timestamps go to the LLM
</li>
<li>
<strong>API keys never persisted in artifacts</strong>:
providerSnapshot stores provider / model / endpoint host
only
</li>
<li>
<strong>All logs redacted</strong>: every
<code>eprintln!</code> path runs through
<code>redact_auth_header</code>, scrubbing
<code>sk-*</code> / <code>Bearer *</code> /
<code>x-api-key</code> / Gemini <code>?key=</code>
</li>
<li>
<strong>HTML reports CSP-enforced</strong>:
<code>default-src 'none'</code> + no script / no
external resources; the in-app preview iframe uses
<code>sandbox="allow-same-origin"</code> (no
<code>allow-scripts</code>), giving two-layer XSS
protection
</li>
<li>
<strong>auto-scenario low-confidence fallback</strong>:
prevents casual/private chats from being misclassified
as "business commitments"
</li>
</ul>
</div>
</div>
</body>
</html>