mulmocast 2.1.37 → 2.1.39

package/lib/actions/bundle.js

@@ -71,7 +71,15 @@ export const mulmoViewerBundle = async (context, options = {}) => {
         const sudioBeats = context.studio.beats[index];
         const { duration, startAt } = sudioBeats;
         // console.log(context.studio.beats[index]);
-        resultJson.push({ text: beat.text, duration, startTime: startAt, endTime: (startAt ?? 0) + (duration ?? 0), audioSources: {}, multiLinguals: {} });
+        resultJson.push({
+            id: beat.id,
+            text: beat.text,
+            duration,
+            startTime: startAt,
+            endTime: (startAt ?? 0) + (duration ?? 0),
+            audioSources: {},
+            multiLinguals: {},
+        });
     });
     // audio
     for (const lang of bundleTargetLang) {

package/lib/methods/mulmo_studio_context.js

@@ -60,19 +60,17 @@ export const MulmoStudioContextMethods = {
         notifyStateChange(context, sessionType, result);
     },
     setBeatSessionState(context, sessionType, index, id, value) {
-        if (!sessionType) {
+        if (!sessionType || !Object.hasOwn(context.sessionState.inBeatSession, sessionType)) {
             return;
         }
         const key = beatId(id, index);
+        const session = context.sessionState.inBeatSession[sessionType];
         if (value) {
-            if (!context.sessionState.inBeatSession[sessionType]) {
-                context.sessionState.inBeatSession[sessionType] = {};
-            }
-            context.sessionState.inBeatSession[sessionType][key] = true;
+            session[key] = true;
         }
         else {
             // NOTE: Setting to false causes the parse error in rebuildStudio in preprocess.ts
-            delete context.sessionState.inBeatSession[sessionType][key];
+            delete session[key];
         }
         notifyBeatStateChange(context, sessionType, key);
     },

package/lib/types/provider2agent.js

@@ -27,7 +27,7 @@ export const provider2TTSAgent = {
         defaultModel: "eleven_multilingual_v2",
         // Models | ElevenLabs Documentation
         // https://elevenlabs.io/docs/models
-        models: ["eleven_multilingual_v2", "eleven_turbo_v2_5", "eleven_turbo_v2", "eleven_flash_v2_5", "eleven_flash_v2"],
+        models: ["eleven_v3", "eleven_multilingual_v2", "eleven_turbo_v2_5", "eleven_turbo_v2", "eleven_flash_v2_5", "eleven_flash_v2"],
         keyName: "ELEVENLABS_API_KEY",
     },
     kotodama: {

package/lib/types/viewer.d.ts

@@ -1,5 +1,6 @@
 import { z } from "zod";
 export declare const mulmoViewerBeatSchema: z.ZodObject<{
+    id: z.ZodOptional<z.ZodString>;
     text: z.ZodOptional<z.ZodString>;
     duration: z.ZodOptional<z.ZodNumber>;
     startTime: z.ZodOptional<z.ZodNumber>;
@@ -16,6 +17,7 @@ export declare const mulmoViewerBeatSchema: z.ZodObject<{
 export type MulmoViewerBeat = z.infer<typeof mulmoViewerBeatSchema>;
 export declare const mulmoViewerDataSchema: z.ZodObject<{
     beats: z.ZodArray<z.ZodObject<{
+        id: z.ZodOptional<z.ZodString>;
         text: z.ZodOptional<z.ZodString>;
         duration: z.ZodOptional<z.ZodNumber>;
         startTime: z.ZodOptional<z.ZodNumber>;

package/lib/types/viewer.js

@@ -1,5 +1,6 @@
 import { z } from "zod";
 export const mulmoViewerBeatSchema = z.object({
+    id: z.string().optional(),
     text: z.string().optional(),
     duration: z.number().optional(),
     startTime: z.number().optional(),

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mulmocast",
-  "version": "2.1.37",
+  "version": "2.1.39",
   "description": "",
   "type": "module",
   "main": "lib/index.node.js",
@@ -76,9 +76,9 @@
   "homepage": "https://github.com/receptron/mulmocast-cli#readme",
   "dependencies": {
     "@google-cloud/text-to-speech": "^6.4.0",
-    "@google/genai": "^1.40.0",
+    "@google/genai": "^1.41.0",
     "@graphai/anthropic_agent": "^2.0.12",
-    "@graphai/browserless_agent": "^2.0.1",
+    "@graphai/browserless_agent": "^2.0.2",
     "@graphai/gemini_agent": "^2.0.5",
     "@graphai/groq_agent": "^2.0.2",
     "@graphai/input_agents": "^1.0.2",
@@ -92,12 +92,12 @@
     "@mozilla/readability": "^0.6.0",
     "@tavily/core": "^0.5.11",
     "archiver": "^7.0.1",
-    "clipboardy": "^5.2.1",
-    "dotenv": "^17.2.4",
+    "clipboardy": "^5.3.0",
+    "dotenv": "^17.3.1",
     "fluent-ffmpeg": "^2.1.3",
     "graphai": "^2.0.16",
     "jsdom": "^28.0.0",
-    "marked": "^17.0.1",
+    "marked": "^17.0.2",
     "mulmocast-vision": "^1.0.8",
     "ora": "^9.3.0",
     "puppeteer": "^24.37.2",
@@ -117,12 +117,12 @@
     "eslint": "^10.0.0",
     "eslint-config-prettier": "^10.1.8",
     "eslint-plugin-prettier": "^5.5.5",
-    "eslint-plugin-sonarjs": "^3.0.6",
+    "eslint-plugin-sonarjs": "^3.0.7",
     "globals": "^17.3.0",
     "prettier": "^3.8.1",
     "tsx": "^4.21.0",
     "typescript": "^5.9.3",
-    "typescript-eslint": "^8.54.0"
+    "typescript-eslint": "^8.55.0"
   },
   "engines": {
     "node": ">=22.0.0"

package/scripts/test/test_elevenlabs_models.json

@@ -64,6 +64,22 @@
           "en": "Flash V2"
         }
       },
+      "V3": {
+        "provider": "elevenlabs",
+        "voiceId": "1SM7GgM6IMuvQlz2BwM3",
+        "model": "eleven_v3",
+        "displayName": {
+          "en": "Eleven V3 (Mark)"
+        }
+      },
+      "V3Japanese": {
+        "provider": "elevenlabs",
+        "voiceId": "KtSs8OSniRPofXnr5PeA",
+        "model": "eleven_v3",
+        "displayName": {
+          "en": "V3 Japanese (Sumire)"
+        }
+      },
       "MultilingualJapanese": {
         "provider": "elevenlabs",
         "voiceId": "Mv8AjrYZCBkdsmDHNwcB",
@@ -157,6 +173,28 @@
         }
       }
     },
+    {
+      "speaker": "V3",
+      "text": "Testing Eleven v3, the latest and most advanced model with human-like expressiveness and high emotional range across over 70 languages.",
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Eleven V3",
+          "subtitle": "Latest model, 70+ languages"
+        }
+      }
+    },
+    {
+      "speaker": "V3Japanese",
+      "text": "Eleven V3モデルの日本語テストです。最新の音声合成モデルで、70以上の言語に対応し、豊かな感情表現が可能です。",
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Eleven V3日本語テスト",
+          "subtitle": "最新モデル、70+言語対応"
+        }
+      }
+    },
     {
       "speaker": "MultilingualJapanese",
       "text": "こんにちは。これは多言語モデルV2の日本語テストです。29言語に対応し、豊かな感情表現を持つプロフェッショナル品質のモデルです。",