feat: add OCR engine profiles

CHANGELOG.md

@@ -8,6 +8,9 @@ Follows [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) · versioned wi
 ## [Unreleased]
 
 ### Changed
+- **OCR engine category.** Engines now includes local OCR/document-parser profiles for KDL Frontier Parser nano and
+  Baidu Unlimited-OCR, with custom OCR profiles, connection probes, page-image API guidance, and future Magazine
+  workflow integration documented in the TODO roadmap.
 - **Initial audiobook casts preserve dialogue boundaries.** The final cast
   normalization now merges adjacent narration only; same-speaker dialogue
   rows remain separate so Run 3 segmentation errors stay visible and do not

TODO.md

@@ -283,6 +283,36 @@ codebase; benefit is relative user/reliability impact.
 - [ ] **P2 / M–L — Medium–high benefit — Enrich character-form UX.** Add richer inline form editing and per-form
   bulk-generation controls while preserving one character ID and one voice.
 
+## Magazine reader roadmap
+
+Magazines are a separate Studio document type: they use a narrator and layout-aware article extraction, not character
+casting. The original PDF remains the visual source of truth while approved articles become linked, reflowable HTML.
+
+### P0 — Magazine foundation
+
+- [ ] Add a Magazine project type beside Books and Theater Plays.
+- [ ] Add PDF/JPG/PNG import with page previews and safe local storage.
+- [x] Add OCR-engine selection from Engines > OCR & Document Parsing, with page-image input and structured output.
+- [ ] Detect contents/index pages and propose article titles, authors, and page ranges.
+- [ ] Build a review screen where users approve or correct article boundaries before TTS.
+- [ ] Preserve source coordinates for every extracted block, image, caption, and page reference.
+
+### P1 — Magazine listening experience
+
+- [ ] Generate clean article HTML with headings, captions, figures, sidebars, and page links.
+- [ ] Generate narrator audio per article with resume/progress and a play-all queue.
+- [ ] Synchronize paragraph/sentence highlighting with playback and keep PDF and article views linked.
+- [ ] Show article images or page crops during playback, with optional caption reading.
+- [ ] Add confidence warnings, extraction diffs, and manual text correction before audio generation.
+
+### P2 — Advanced magazine intelligence
+
+- [ ] Improve cross-page continuation detection and automatic sidebar/advertisement exclusion.
+- [ ] Add image descriptions, chart/table handling, and optional spoken figure summaries.
+- [ ] Add word-level karaoke timing and EPUB 3 Media Overlay export.
+- [ ] Add mobile/driving mode with large controls, automatic article advance, and hands-free resume.
+- [ ] Add OCR-engine benchmark comparisons and per-page fallback selection.
+
 Three full regression runs of *Der Sturm* against the hand-corrected reference.
 Everything below is measured, not estimated. Artefacts live in
 `config/regression_snapshots/der_sturm/` (`*_manual.json` are the human

VERSION

@@ -1 +1 @@
-1.25.1
+1.25.2

routes/docker.py

@@ -43,6 +43,8 @@ _LOCAL_CONTAINER_DEFS: list[dict] = [
     {"name": "voicebox",                      "label": "Voicebox",                   "role": "llm", "port": 17493, "stack": "voicebox", "image": "local", "repo": "", "description": "Companion voice app/control plane. Not used as a normal TTS engine unless wrapped as an API."},
     {"name": "stt-whisperx-blackwell",        "label": "WhisperX · faster-whisper (Blackwell GPU)", "role": "stt", "port": 8003, "stack": "whisperx-blackwell", "image": "local", "repo": "https://github.com/m-bain/whisperX", "description": "WhisperX transcription/alignment service using faster-whisper on the Blackwell GPU. Use as custom STT if its API is OpenAI-compatible or /transcribe."},
     {"name": "xtts-dgx-spark",                "label": "XTTS v2",                    "role": "tts", "port": 8005, "stack": "xtts_v2", "image": "local", "repo": "https://github.com/daswer123/xtts-api-server", "description": "XTTS v2 OpenAI-compatible speaker endpoint."},
+    {"name": "ocr-kdl-frontier-parser-nano",  "label": "KDL Frontier Parser nano",    "role": "ocr", "port": 8099, "stack": "", "image": "local", "repo": "https://huggingface.co/KDLAI/KDL-Frontier-Parser-nano", "description": "Local 1.2B document parser for layout detection, OCR, tables, figures, and reading order."},
+    {"name": "ocr-baidu-unlimited-ocr",       "label": "Baidu Unlimited-OCR",          "role": "ocr", "port": 8098, "stack": "", "image": "local", "repo": "https://github.com/baidu/Unlimited-OCR", "description": "Local long-document OCR and layout parser with single-image and multi-page/PDF workflows."},
 ]
 
 
@@ -153,6 +155,8 @@ async def probe_url(url: str, type: str = "", provider_api_key: str = Header("",
         checks = [("/health", None), ("/v1/models", "data"), ("/v1/audio/transcriptions", None)]
     elif type == "tts":
         checks = [("/health", None), ("/v1/health", None), ("/v1/audio/voices", None), ("/speakers", None), ("/voices", None)]
+    elif type == "ocr":
+        checks = [("/health", None), ("/v1/models", "data"), ("/v1/chat/completions", None), ("/parse", None), ("/ocr", None)]
     elif type == "comfyui":
         checks = [("/system_stats", "system")]
     else:

static/index.html

@@ -10,7 +10,7 @@
   <meta name="format-detection" content="telephone=no">
   <meta name="color-scheme" content="light dark">
   <meta name="theme-color" content="#2563EB">
-<meta name="app-version" content="1.25.1">
+<meta name="app-version" content="1.25.2">
   <link rel="manifest" href="/manifest.webmanifest">
   <link rel="icon" href="/static/icon.svg" type="image/svg+xml">
   <link rel="apple-touch-icon" href="/static/icon.svg">
@@ -27,12 +27,12 @@
 
   <!-- ── Core styles (local — no CDN dependency for first paint) ────────── -->
   <link rel="stylesheet" href="/static/vendor/mdi/materialdesignicons.min.css">
-<link rel="stylesheet" href="/static/style.css?v=1.25.1">
-<link rel="stylesheet" href="/static/css/resource-diagnostics.css?v=1.25.1">
-<link rel="stylesheet" href="/static/css/background-jobs.css?v=1.25.1">
-<link rel="stylesheet" href="/static/css/voice-design.css?v=1.25.1">
-<link rel="stylesheet" href="/static/css/rehearser.css?v=1.25.1">
-<link rel="stylesheet" href="/static/css/audiobook.css?v=1.25.1">
+<link rel="stylesheet" href="/static/style.css?v=1.25.2">
+<link rel="stylesheet" href="/static/css/resource-diagnostics.css?v=1.25.2">
+<link rel="stylesheet" href="/static/css/background-jobs.css?v=1.25.2">
+<link rel="stylesheet" href="/static/css/voice-design.css?v=1.25.2">
+<link rel="stylesheet" href="/static/css/rehearser.css?v=1.25.2">
+<link rel="stylesheet" href="/static/css/audiobook.css?v=1.25.2">
 
 
   <!-- ── Flag icons — non-blocking (loaded async, icons appear after JS) ── -->
@@ -220,6 +220,7 @@
           <div class="nav-tree-item"           data-engines-cat="stt" onclick="navEnginesCat('stt')"><span class="mdi mdi-ear-hearing"></span> Speech to Text</div>
           <div class="nav-tree-item"           data-engines-cat="tts" onclick="navEnginesCat('tts')"><span class="mdi mdi-account-voice"></span> Text to Speech</div>
           <div class="nav-tree-item"           data-engines-cat="image" onclick="navEnginesCat('image')"><span class="mdi mdi-image-outline"></span> Image Generation</div>
+          <div class="nav-tree-item"           data-engines-cat="ocr" onclick="navEnginesCat('ocr')"><span class="mdi mdi-file-document-multiple-outline"></span> OCR &amp; Document Parsing</div>
         </div>
         <div class="nav-tree-head nav-subhead" id="nav-integrations-head" onclick="toggleNavTree('nav-integrations-tree','nav-integrations-chevron')">
           <span class="nav-icon"><span class="mdi mdi-transit-connection-variant"></span></span>
@@ -397,7 +398,7 @@ window.toggleNavTree = function(treeId, chevronId) {
 </script>
 
 <!-- loader.js: fetches sections → loads JS modules → removes skeleton -->
-<script src="/static/loader.js?v=1.25.1"></script>
+<script src="/static/loader.js?v=1.25.2"></script>
 
 </body>
 </html>

static/js/engines.js

@@ -392,11 +392,13 @@ async function loadLocalContainers() {
   const gridTts = $('dc-grid-tts');
   const gridStt = $('dc-grid-stt');
   const gridLlm = $('dc-grid-llm');
-  if (!gridTts && !gridStt && !gridLlm) return;
+  const gridOcr = $('dc-grid-ocr');
+  if (!gridTts && !gridStt && !gridLlm && !gridOcr) return;
   const skel = _dcSkel(2);
   if (gridTts && !gridTts.querySelector('.llm-local-card')) gridTts.innerHTML = skel;
   if (gridStt && !gridStt.querySelector('.llm-local-card')) gridStt.innerHTML = skel;
   if (gridLlm && !gridLlm.querySelector('.llm-local-card')) gridLlm.innerHTML = skel;
+  if (gridOcr && !gridOcr.querySelector('.llm-local-card')) gridOcr.innerHTML = skel;
   const custom = loadCustomEngineCards();
   try {
     const d = await fetchLocalContainersWithRetry();
@@ -417,6 +419,7 @@ async function loadLocalContainers() {
     if (gridTts) gridTts.innerHTML = err;
     if (gridStt) gridStt.innerHTML = err;
     if (gridLlm) gridLlm.innerHTML = err;
+    if (gridOcr) gridOcr.innerHTML = err;
     document.querySelectorAll('.dc-retry-inline').forEach(btn => btn.addEventListener('click', loadLocalContainers));
   }
 }
@@ -426,9 +429,10 @@ function renderLocalContainers(containers) {
   const gridTts = $('dc-grid-tts');
   const gridStt = $('dc-grid-stt');
   const gridLlm = $('dc-grid-llm');
-  if (!gridTts && !gridStt && !gridLlm) return;
+  const gridOcr = $('dc-grid-ocr');
+  if (!gridTts && !gridStt && !gridLlm && !gridOcr) return;
 
-  const ROLE_LABEL = { tts: 'TTS', stt: 'STT', 'stt+tts': 'STT · TTS', llm: 'LLM' };
+  const ROLE_LABEL = { tts: 'TTS', stt: 'STT', 'stt+tts': 'STT · TTS', llm: 'LLM', ocr: 'OCR' };
   const DC_ICONS = {
     'tts-faster-qwen3-tts-voiceclone':  '<span class="mdi mdi-speaker-outline"></span>',
     'tts-faster-qwen3-tts-voicedesign': '<span class="mdi mdi-auto-fix"></span>',
@@ -448,6 +452,8 @@ function renderLocalContainers(containers) {
     'nemotron-asr-streaming':        '<span class="mdi mdi-waveform"></span>',
     'vibevoice-msft-tts':            '<span class="mdi mdi-account-voice"></span>',
     'xtts-dgx-spark':                '<span class="mdi mdi-account-music-outline"></span>',
+    'ocr-kdl-frontier-parser-nano':  '<span class="mdi mdi-file-document-outline"></span>',
+    'ocr-baidu-unlimited-ocr':       '<span class="mdi mdi-file-document-multiple-outline"></span>',
     'litellm':                       '<span class="mdi mdi-router-network"></span>',
     'llama.cpp':                     '<span class="mdi mdi-chip"></span>',
     'ollama':                        '<span class="mdi mdi-circle-outline"></span>',
@@ -473,8 +479,10 @@ function renderLocalContainers(containers) {
     'llama.cpp':                     [[_t(),'local'],[_l(),'model dependent'],[_q(),'OpenAI API'],[_m(),'model dependent']],
     'ollama':                        [[_t(),'local'],[_l(),'model dependent'],[_q(),'OpenAI API'],[_m(),'model dependent']],
     'llama-swap':                    [[_t(),'swap'],[_l(),'warm after load'],[_q(),'OpenAI API'],[_m(),'model dependent']],
+    'ocr-kdl-frontier-parser-nano': [[_t(),'local'],[_l(),'layout-aware'],[_q(),'1.2B parser'],[_m(),'model dependent']],
+    'ocr-baidu-unlimited-ocr':      [[_t(),'local'],[_l(),'multi-page'],[_q(),'long-horizon'],[_m(),'model dependent']],
   };
-  const roleIcon = { tts: '<span class="mdi mdi-speaker-outline"></span>', stt: '<span class="mdi mdi-microphone-variant"></span>', 'stt+tts': '<span class="mdi mdi-swap-horizontal"></span>', llm: '<span class="mdi mdi-robot-outline"></span>' };
+  const roleIcon = { tts: '<span class="mdi mdi-speaker-outline"></span>', stt: '<span class="mdi mdi-microphone-variant"></span>', 'stt+tts': '<span class="mdi mdi-swap-horizontal"></span>', llm: '<span class="mdi mdi-robot-outline"></span>', ocr: '<span class="mdi mdi-file-document-outline"></span>' };
 
   function buildCardHtml(c) {
     const isCustom = !!c.isCustom;
@@ -577,6 +585,10 @@ function renderLocalContainers(containers) {
       snippetCode = `curl -s "${snippetBaseUrl}/v1/audio/speech" \\\n  -H "Authorization: Bearer dummy" \\\n  -H "Content-Type: application/json" \\\n  -d '{"model":"tts-1","voice":"default","input":"Hello world","response_format":"wav"}' \\\n  --output test.wav`;
     } else if (c.role === 'stt') {
       snippetCode = `curl -s "${snippetBaseUrl}/v1/audio/transcriptions" \\\n  -F "file=@audio.wav" \\\n  -F "model=whisper-1"`;
+    } else if (c.role === 'ocr') {
+      snippetCode = `curl -s "${snippetBaseUrl}/v1/chat/completions" \\
+  -H "Content-Type: application/json" \\
+  -d '{"model":"${c.name}","messages":[{"role":"user","content":[{"type":"text","text":"\\nText Recognition:"},{"type":"image_url","image_url":{"url":"data:image/png;base64,..."}}]}]}'`;
     } else {
       snippetCode = `curl -s "${snippetBaseUrl}/v1/chat/completions" \\\n  -H "Authorization: Bearer dummy" \\\n  -H "Content-Type: application/json" \\\n  -d '{"model":"default","messages":[{"role":"user","content":"Hello!"}]}'`;
     }
@@ -605,12 +617,14 @@ function renderLocalContainers(containers) {
   const isTts = c => c.role === 'tts';
   const isStt = c => c.role === 'stt' || c.role === 'stt+tts';
   const isLlm = c => c.role === 'llm';
+  const isOcr = c => c.role === 'ocr';
 
   if (gridTts) gridTts.innerHTML = containers.filter(isTts).map(buildCardHtml).join('');
   if (gridStt) gridStt.innerHTML = containers.filter(isStt).map(buildCardHtml).join('');
   if (gridLlm) gridLlm.innerHTML = containers.filter(isLlm).map(buildCardHtml).join('');
+  if (gridOcr) gridOcr.innerHTML = containers.filter(isOcr).map(buildCardHtml).join('');
 
-  [gridTts, gridStt, gridLlm].forEach(grid => {
+  [gridTts, gridStt, gridLlm, gridOcr].forEach(grid => {
     if (!grid) return;
 
     // Stop / Start / Restart docker containers

static/sections/s-llms.html

@@ -43,6 +43,18 @@
             <p class="llm-sec-note">Used to generate character profile pictures. Pick the active provider below and add its API key.</p>
           </div>
         </div>
+
+        <div class="engines-head-context header-cat-block" data-engines-cat="ocr">
+          <div class="engines-context-copy">
+            <h3 class="llm-sec-title">OCR &amp; Document Parsing</h3>
+            <p class="llm-sec-note">Configure local OCR and layout-aware document parsers for scanned PDFs, images, and future Magazine projects.</p>
+          </div>
+          <div class="engines-context-actions">
+            <button class="btn-secondary" type="button" onclick="openAddEngineDialog('ocr')"><span class="mdi mdi-plus"></span> Add Custom</button>
+            <button class="btn-secondary dc-refresh-btn" type="button"><span class="mdi mdi-refresh"></span> Refresh</button>
+            <span class="llm-free-badge llm-free-local">100% Local</span>
+          </div>
+        </div>
       </div>
 
       <!-- ── Language Models sub-page ──────────────────────────── -->
@@ -809,6 +821,7 @@
               <option value="tts">Text to Speech (TTS)</option>
               <option value="stt">Speech to Text (STT)</option>
               <option value="llm">Language Model (LLM)</option>
+              <option value="ocr">OCR &amp; Document Parser</option>
             </select>
           </div>
           <div class="add-engine-field">