mulmocast 2.0.2 → 2.0.3

package/README.md

@@ -440,6 +440,28 @@ Options:
       --image_width        Image width (e.g., 400px, 50%, auto)         [string]
 ```
 
+```
+mulmo bundle <file>
+
+Generate bundle files
+
+Positionals:
+  file  Mulmo Script File                                    [string] [required]
+
+Options:
+      --version            Show version number                         [boolean]
+  -v, --verbose            verbose log     [boolean] [required] [default: false]
+  -h, --help               Show help                                   [boolean]
+  -o, --outdir             output dir                                   [string]
+  -b, --basedir            base dir                                     [string]
+  -l, --lang               target language
+  [string] [choices: "en", "ja", "fr", "es", "de", "zh-CN", "zh-TW", "ko", "it",
+                                                               "pt", "ar", "hi"]
+  -f, --force              Force regenerate           [boolean] [default: false]
+      --backup             create backup media file   [boolean] [default: false]
+  -p, --presentationStyle  Presentation Style                           [string]
+```
+
 ```
 mulmo tool <command>
 

package/assets/templates/ghibli_comic_strips.json

@@ -13,7 +13,7 @@
     },
     "imageParams": {
       "provider": "google",
-      "model": "gemini-2.5-flash-image-preview",
+      "model": "gemini-2.5-flash-image",
       "style": "<style>Ghibli style multi-panel comic strips in landscape mode. Use speech bubbles with short, natural dialogue (1–6 words). Keep text minimal, like real comics. Let the art convey the story and emotions. Use the input image as the presenter.</style>",
       "images": {
         "presenter": {

package/lib/actions/bundle.js

@@ -58,8 +58,11 @@ export const mulmoViewerBundle = async (context) => {
     const zipper = new ZipBuilder(path.resolve(dir, zipFileName));
     // text
     const resultJson = [];
-    context.studio.script.beats.forEach((beat) => {
-        resultJson.push({ text: beat.text, duration: beat.duration, audioSources: {}, multiLinguals: {} });
+    context.studio.script.beats.forEach((beat, index) => {
+        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: {} });
     });
     // audio
     for (const lang of bundleTargetLang) {

package/lib/agents/image_genai_agent.js

@@ -71,7 +71,7 @@ export const imageGenAIAgent = async ({ namedInputs, params, config, }) => {
     }
     try {
         const ai = new GoogleGenAI({ apiKey });
-        if (model === "gemini-2.5-flash-image-preview") {
+        if (model === "gemini-2.5-flash-image" || model === "gemini-3-pro-image-preview") {
             const contents = getGeminiContents(prompt, aspectRatio, referenceImages);
             const response = await ai.models.generateContent({ model, contents });
             return geminiFlashResult(response);

package/lib/agents/image_openai_agent.js

@@ -17,7 +17,7 @@ export const imageOpenaiAgent = async ({ namedInputs, params, config, }) => {
     const model = params.model ?? provider2ImageAgent["openai"].defaultModel;
     const openai = new OpenAI({ apiKey, baseURL });
     const size = (() => {
-        if (model === "gpt-image-1") {
+        if (model === "gpt-image-1" || model === "gpt-image-1-mini") {
             if (canvasSize.width > canvasSize.height) {
                 return "1536x1024";
             }
@@ -46,7 +46,7 @@ export const imageOpenaiAgent = async ({ namedInputs, params, config, }) => {
         n: 1,
         size,
     };
-    if (model === "gpt-image-1") {
+    if (model === "gpt-image-1" || model === "gpt-image-1-mini") {
         imageOptions.moderation = moderation || "auto";
         imageOptions.background = "opaque";
         if (quality) {

package/lib/agents/movie_genai_agent.js

@@ -43,7 +43,7 @@ export const movieGenAIAgent = async ({ namedInputs, params, config, }) => {
             },
             image: undefined,
         };
-        if (model === "veo-3.0-generate-preview") {
+        if (model === "veo-3.0-generate-001" || model === "veo-3.1-generate-preview") {
             payload.config.durationSeconds = undefined;
         }
         if (imagePath) {

package/lib/data/promptTemplates.d.ts

@@ -39,6 +39,10 @@ export declare const promptTemplates: ({
             provider?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -58,7 +62,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -181,6 +184,10 @@ export declare const promptTemplates: ({
             style?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -200,7 +207,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -233,6 +239,10 @@ export declare const promptTemplates: ({
             provider?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -252,7 +262,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -298,6 +307,10 @@ export declare const promptTemplates: ({
             provider?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -317,7 +330,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -363,6 +375,10 @@ export declare const promptTemplates: ({
             provider: string;
             style: string;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -382,7 +398,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -434,6 +449,10 @@ export declare const promptTemplates: ({
             provider?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -453,7 +472,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -486,6 +504,10 @@ export declare const promptTemplates: ({
             provider?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -515,7 +537,6 @@ export declare const promptTemplates: ({
                 Presenter?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -548,6 +569,10 @@ export declare const promptTemplates: ({
             provider?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -567,7 +592,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -613,6 +637,10 @@ export declare const promptTemplates: ({
             provider?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -632,7 +660,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;
@@ -668,6 +695,10 @@ export declare const promptTemplates: ({
             provider?: undefined;
             model?: undefined;
         };
+        movieParams: {
+            provider: string;
+            model?: undefined;
+        };
         soundEffectParams: {
             provider: string;
         };
@@ -687,7 +718,6 @@ export declare const promptTemplates: ({
                 Teacher?: undefined;
             };
         };
-        movieParams?: undefined;
     };
     scriptName: string;
     systemPrompt: string;

package/lib/data/promptTemplates.js

@@ -32,6 +32,9 @@ export const promptTemplates = [
                 },
                 style: "<style>AKIRA aesthetic.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -149,6 +152,9 @@ export const promptTemplates = [
                 images: {},
                 provider: "openai",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -192,6 +198,9 @@ export const promptTemplates = [
             imageParams: {
                 style: "A hand-drawn style illustration with a warm, nostalgic atmosphere. The background is rich with natural scenery—lush forests, cloudy skies, and traditional Japanese architecture. Characters have expressive eyes, soft facial features, and are portrayed with gentle lighting and subtle shading. The color palette is muted yet vivid, using earthy tones and watercolor-like textures. The overall scene feels magical and peaceful, with a sense of quiet wonder and emotional depth, reminiscent of classic 1980s and 1990s Japanese animation.",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -242,6 +251,9 @@ export const promptTemplates = [
             imageParams: {
                 style: "<style>A multi panel comic strips. 1990s American workplace humor. Clean, minimalist line art with muted colors. One character is a nerdy office worker with glasses</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -285,6 +297,9 @@ export const promptTemplates = [
             imageParams: {
                 style: "<style>Photo realistic and cinematic. Let the art convey the story and emotions without text. Use the image for the aspect ratio</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -337,6 +352,9 @@ export const promptTemplates = [
                 },
                 style: "<style>Dragon Ball/Dr. Slump aesthetic.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -389,6 +407,9 @@ export const promptTemplates = [
                 },
                 style: "<style>Ghibli style</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -439,10 +460,13 @@ export const promptTemplates = [
                         type: "image",
                     },
                 },
-                model: "gemini-2.5-flash-image-preview",
+                model: "gemini-2.5-flash-image",
                 provider: "google",
                 style: "<style>Ghibli style multi-panel comic strips in landscape mode. Use speech bubbles with short, natural dialogue (1–6 words). Keep text minimal, like real comics. Let the art convey the story and emotions. Use the input image as the presenter.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -502,6 +526,9 @@ export const promptTemplates = [
                 },
                 style: "<style>Ghost in the shell aesthetic.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -546,6 +573,9 @@ export const promptTemplates = [
                 images: {},
                 provider: "openai",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -590,6 +620,9 @@ export const promptTemplates = [
                 images: {},
                 provider: "openai",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -642,6 +675,9 @@ export const promptTemplates = [
                 },
                 style: "<style>One Piece aesthetic.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -694,6 +730,9 @@ export const promptTemplates = [
                 },
                 style: "<style>Photo realistic, cinematic.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -746,6 +785,9 @@ export const promptTemplates = [
                 },
                 style: "<style>Photo realistic, cinematic.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -789,6 +831,9 @@ export const promptTemplates = [
             imageParams: {
                 style: "<style>Ghibli style. Student (Taro) is a young teenager with a dark short hair with glasses. Teacher is a middle-aged man with grey hair and moustache.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -845,6 +890,9 @@ export const promptTemplates = [
             imageParams: {
                 style: "<style>Photo realistic, cinematic.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -898,6 +946,9 @@ export const promptTemplates = [
                 },
                 style: "<style>A dreamy, hyper-detailed anime style that blends photorealistic backgrounds with vibrant, saturated colors. The skies are often filled with luminous clouds, dazzling sunsets, or star-filled nights, rendered with a glowing, almost ethereal quality. Urban landscapes and rural scenery are meticulously illustrated, with attention to tiny details like reflections in puddles, neon lights, or the texture of grass swaying in the wind. Characters are drawn with soft, expressive features, standing out against the breathtaking environments, creating a sense of emotional depth and lyrical atmosphere. The overall mood is cinematic, romantic, and filled with a sense of fleeting beauty and longing.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },
@@ -944,6 +995,9 @@ export const promptTemplates = [
             imageParams: {
                 style: "<style>Photo realistic, cinematic.</style>",
             },
+            movieParams: {
+                provider: "replicate",
+            },
             soundEffectParams: {
                 provider: "replicate",
             },

package/lib/data/templateDataSet.js

@@ -41,7 +41,7 @@ export const templateDataSet = {
         "```",
     ghibli_comic_strips: "Another AI will generate comic strips for each beat based on the text description of that beat. Mention the reference in one of beats, if it exists. Use the JSON below as a template.\n" +
         "```JSON\n" +
-        `{"$mulmocast":{"version":"1.1","credit":"closing"},"title":"[TITLE: Brief, engaging title for the topic]","lang":"en","references":[{"url":"[SOURCE_URL: URL of the source material]","title":"[SOURCE_TITLE: Title of the referenced article, or paper]","type":"[SOURCE_TYPE: article, paper]"}],"beats":[{"text":"[OPENING_BEAT: Introduce the topic with a hook. Reference the source material and set up why this topic matters. Usually 2-3 sentences that grab attention and provide context.]"},{"text":"[MAIN_CONCEPT: Define or explain the core concept/idea. This should be the central focus of your narrative. Keep it clear and accessible.]"},{"text":"[SUPPORTING_DETAIL_1: Additional context, examples, or elaboration that helps illustrate the main concept. This could include how it works, why it's important, or real-world applications.]"},{"text":"[SUPPORTING_DETAIL_2: Continue with more examples, deeper explanation, or different aspects of the topic if needed.]"},{"text":"[ADDITIONAL_BEATS: Add more beats as necessary to fully explore the topic. Complex topics may require 6-10+ beats to cover adequately. Each beat should advance the narrative or provide valuable information.]"},{"text":"[CONCLUSION/IMPACT: Wrap up with the significance, implications, or key takeaway. Help the audience understand why this matters to them.]"}],"canvasSize":{"width":1536,"height":1024},"imageParams":{"provider":"google","model":"gemini-2.5-flash-image-preview","style":"<style>Ghibli style multi-panel comic strips in landscape mode. Use speech bubbles with short, natural dialogue (1–6 words). Keep text minimal, like real comics. Let the art convey the story and emotions. Use the input image as the presenter.</style>","images":{"presenter":{"type":"image","source":{"kind":"url","url":"https://raw.githubusercontent.com/receptron/mulmocast-media/refs/heads/main/characters/ghibli_presenter.png"}}}}}\n` +
+        `{"$mulmocast":{"version":"1.1","credit":"closing"},"title":"[TITLE: Brief, engaging title for the topic]","lang":"en","references":[{"url":"[SOURCE_URL: URL of the source material]","title":"[SOURCE_TITLE: Title of the referenced article, or paper]","type":"[SOURCE_TYPE: article, paper]"}],"beats":[{"text":"[OPENING_BEAT: Introduce the topic with a hook. Reference the source material and set up why this topic matters. Usually 2-3 sentences that grab attention and provide context.]"},{"text":"[MAIN_CONCEPT: Define or explain the core concept/idea. This should be the central focus of your narrative. Keep it clear and accessible.]"},{"text":"[SUPPORTING_DETAIL_1: Additional context, examples, or elaboration that helps illustrate the main concept. This could include how it works, why it's important, or real-world applications.]"},{"text":"[SUPPORTING_DETAIL_2: Continue with more examples, deeper explanation, or different aspects of the topic if needed.]"},{"text":"[ADDITIONAL_BEATS: Add more beats as necessary to fully explore the topic. Complex topics may require 6-10+ beats to cover adequately. Each beat should advance the narrative or provide valuable information.]"},{"text":"[CONCLUSION/IMPACT: Wrap up with the significance, implications, or key takeaway. Help the audience understand why this matters to them.]"}],"canvasSize":{"width":1536,"height":1024},"imageParams":{"provider":"google","model":"gemini-2.5-flash-image","style":"<style>Ghibli style multi-panel comic strips in landscape mode. Use speech bubbles with short, natural dialogue (1–6 words). Keep text minimal, like real comics. Let the art convey the story and emotions. Use the input image as the presenter.</style>","images":{"presenter":{"type":"image","source":{"kind":"url","url":"https://raw.githubusercontent.com/receptron/mulmocast-media/refs/heads/main/characters/ghibli_presenter.png"}}}}}\n` +
         "```",
     ghost_comic: "Another AI will generate images for each beat based on the image prompt of that beat. Mention the reference in one of beats, if it exists. Use the JSON below as a template.\n" +
         "```JSON\n" +

package/lib/methods/mulmo_presentation_style.js

@@ -91,7 +91,7 @@ export const MulmoPresentationStyleMethods = {
     },
     getMovieAgentInfo(presentationStyle, beat) {
         const movieParams = { ...presentationStyle.movieParams, ...beat?.movieParams };
-        const movieProvider = text2MovieProviderSchema.parse(movieParams?.provider);
+        const movieProvider = text2MovieProviderSchema.parse(movieParams?.provider ?? defaultProviders.text2movie);
         const agentInfo = provider2MovieAgent[movieProvider];
         return {
             agent: agentInfo.agentName,

package/lib/types/schema.d.ts

@@ -453,9 +453,9 @@ export declare const htmlPromptParamsSchema: z.ZodObject<{
     data: z.ZodOptional<z.ZodAny>;
     images: z.ZodOptional<z.ZodRecord<z.ZodString, z.ZodAny>>;
 }, z.core.$strict>;
-export declare const text2MovieProviderSchema: z.ZodDefault<z.ZodEnum<{
+export declare const text2MovieProviderSchema: z.ZodEnum<{
     [x: string]: string;
-}>>;
+}>;
 export declare const text2SoundEffectProviderSchema: z.ZodDefault<z.ZodEnum<{
     [x: string]: string;
 }>>;
@@ -601,9 +601,9 @@ export declare const mulmoBeatSchema: z.ZodObject<{
         movieVolume: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
     }, z.core.$strict>>;
     movieParams: z.ZodOptional<z.ZodObject<{
-        provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+        provider: z.ZodOptional<z.ZodEnum<{
             [x: string]: string;
-        }>>>;
+        }>>;
         model: z.ZodOptional<z.ZodString>;
         fillOption: z.ZodOptional<z.ZodObject<{
             style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -681,9 +681,9 @@ export declare const mulmoTransitionSchema: z.ZodObject<{
     duration: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
 }, z.core.$strip>;
 export declare const mulmoMovieParamsSchema: z.ZodObject<{
-    provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+    provider: z.ZodOptional<z.ZodEnum<{
         [x: string]: string;
-    }>>>;
+    }>>;
     model: z.ZodOptional<z.ZodString>;
     fillOption: z.ZodOptional<z.ZodObject<{
         style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -761,10 +761,10 @@ export declare const mulmoPresentationStyleSchema: z.ZodObject<{
             prompt: z.ZodString;
         }, z.core.$strict>]>>>;
     }, z.core.$strict>>>;
-    movieParams: z.ZodOptional<z.ZodObject<{
-        provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+    movieParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
+        provider: z.ZodOptional<z.ZodEnum<{
             [x: string]: string;
-        }>>>;
+        }>>;
         model: z.ZodOptional<z.ZodString>;
         fillOption: z.ZodOptional<z.ZodObject<{
             style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -779,7 +779,7 @@ export declare const mulmoPresentationStyleSchema: z.ZodObject<{
             }>;
             duration: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
         }, z.core.$strip>>;
-    }, z.core.$strict>>;
+    }, z.core.$strict>>>;
     soundEffectParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
         provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
             [x: string]: string;
@@ -897,10 +897,10 @@ export declare const mulmoScriptSchema: z.ZodObject<{
             prompt: z.ZodString;
         }, z.core.$strict>]>>>;
     }, z.core.$strict>>>;
-    movieParams: z.ZodOptional<z.ZodObject<{
-        provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+    movieParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
+        provider: z.ZodOptional<z.ZodEnum<{
             [x: string]: string;
-        }>>>;
+        }>>;
         model: z.ZodOptional<z.ZodString>;
         fillOption: z.ZodOptional<z.ZodObject<{
             style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -915,7 +915,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
             }>;
             duration: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
         }, z.core.$strip>>;
-    }, z.core.$strict>>;
+    }, z.core.$strict>>>;
     soundEffectParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
         provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
             [x: string]: string;
@@ -1105,9 +1105,9 @@ export declare const mulmoScriptSchema: z.ZodObject<{
             movieVolume: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
         }, z.core.$strict>>;
         movieParams: z.ZodOptional<z.ZodObject<{
-            provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+            provider: z.ZodOptional<z.ZodEnum<{
                 [x: string]: string;
-            }>>>;
+            }>>;
             model: z.ZodOptional<z.ZodString>;
             fillOption: z.ZodOptional<z.ZodObject<{
                 style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -1309,10 +1309,10 @@ export declare const mulmoStudioSchema: z.ZodObject<{
                 prompt: z.ZodString;
             }, z.core.$strict>]>>>;
         }, z.core.$strict>>>;
-        movieParams: z.ZodOptional<z.ZodObject<{
-            provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+        movieParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
+            provider: z.ZodOptional<z.ZodEnum<{
                 [x: string]: string;
-            }>>>;
+            }>>;
             model: z.ZodOptional<z.ZodString>;
             fillOption: z.ZodOptional<z.ZodObject<{
                 style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -1327,7 +1327,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
                 }>;
                 duration: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
             }, z.core.$strip>>;
-        }, z.core.$strict>>;
+        }, z.core.$strict>>>;
         soundEffectParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
             provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                 [x: string]: string;
@@ -1517,9 +1517,9 @@ export declare const mulmoStudioSchema: z.ZodObject<{
                 movieVolume: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
             }, z.core.$strict>>;
             movieParams: z.ZodOptional<z.ZodObject<{
-                provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+                provider: z.ZodOptional<z.ZodEnum<{
                     [x: string]: string;
-                }>>>;
+                }>>;
                 model: z.ZodOptional<z.ZodString>;
                 fillOption: z.ZodOptional<z.ZodObject<{
                     style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -1657,10 +1657,10 @@ export declare const mulmoPromptTemplateSchema: z.ZodObject<{
                 prompt: z.ZodString;
             }, z.core.$strict>]>>>;
         }, z.core.$strict>>>;
-        movieParams: z.ZodOptional<z.ZodObject<{
-            provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+        movieParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
+            provider: z.ZodOptional<z.ZodEnum<{
                 [x: string]: string;
-            }>>>;
+            }>>;
             model: z.ZodOptional<z.ZodString>;
             fillOption: z.ZodOptional<z.ZodObject<{
                 style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -1675,7 +1675,7 @@ export declare const mulmoPromptTemplateSchema: z.ZodObject<{
                 }>;
                 duration: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
             }, z.core.$strip>>;
-        }, z.core.$strict>>;
+        }, z.core.$strict>>>;
         soundEffectParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
             provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                 [x: string]: string;
@@ -1787,10 +1787,10 @@ export declare const mulmoPromptTemplateFileSchema: z.ZodObject<{
                 prompt: z.ZodString;
             }, z.core.$strict>]>>>;
         }, z.core.$strict>>>;
-        movieParams: z.ZodOptional<z.ZodObject<{
-            provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
+        movieParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
+            provider: z.ZodOptional<z.ZodEnum<{
                 [x: string]: string;
-            }>>>;
+            }>>;
             model: z.ZodOptional<z.ZodString>;
             fillOption: z.ZodOptional<z.ZodObject<{
                 style: z.ZodDefault<z.ZodOptional<z.ZodEnum<{
@@ -1805,7 +1805,7 @@ export declare const mulmoPromptTemplateFileSchema: z.ZodObject<{
                 }>;
                 duration: z.ZodDefault<z.ZodOptional<z.ZodNumber>>;
             }, z.core.$strip>>;
-        }, z.core.$strict>>;
+        }, z.core.$strict>>>;
         soundEffectParams: z.ZodDefault<z.ZodOptional<z.ZodObject<{
             provider: z.ZodOptional<z.ZodDefault<z.ZodEnum<{
                 [x: string]: string;

package/lib/types/schema.js

@@ -268,7 +268,7 @@ export const htmlPromptParamsSchema = z
     images: z.record(z.string(), z.any()).optional(),
 })
     .strict();
-export const text2MovieProviderSchema = z.enum(Object.keys(provider2MovieAgent)).default(defaultProviders.text2movie);
+export const text2MovieProviderSchema = z.enum(Object.keys(provider2MovieAgent));
 export const text2SoundEffectProviderSchema = z.enum(Object.keys(provider2SoundEffectAgent)).default(defaultProviders.soundEffect);
 export const mulmoSoundEffectParamsSchema = z.object({
     provider: text2SoundEffectProviderSchema.optional(),
@@ -359,7 +359,9 @@ export const mulmoPresentationStyleSchema = z.object({
         provider: defaultProviders.text2image,
         images: {},
     }),
-    movieParams: mulmoMovieParamsSchema.optional(),
+    movieParams: mulmoMovieParamsSchema.optional().default({
+        provider: defaultProviders.text2movie,
+    }),
     soundEffectParams: mulmoSoundEffectParamsSchema.optional().default({
         provider: defaultProviders.soundEffect,
     }),

package/lib/types/type.d.ts

@@ -139,6 +139,8 @@ export type ImageType = "image" | "movie";
 export type MulmoViewerBeat = {
     text?: string;
     duration?: number;
+    startTime?: number;
+    endTime?: number;
     multiLinguals?: Record<string, string>;
     audioSources?: Record<string, string>;
     imageSource?: string;

package/lib/utils/context.d.ts

@@ -56,6 +56,17 @@ export declare const createStudioData: (_mulmoScript: MulmoScript, fileName: str
                 prompt: string;
             }> | undefined;
         };
+        movieParams: {
+            provider?: string | undefined;
+            model?: string | undefined;
+            fillOption?: {
+                style: "aspectFit" | "aspectFill";
+            } | undefined;
+            transition?: {
+                type: "fade" | "slideout_left";
+                duration: number;
+            } | undefined;
+        };
         soundEffectParams: {
             provider?: string | undefined;
             model?: string | undefined;
@@ -252,17 +263,6 @@ export declare const createStudioData: (_mulmoScript: MulmoScript, fileName: str
             enableLipSync?: boolean | undefined;
             hidden?: boolean | undefined;
         }[];
-        movieParams?: {
-            provider?: string | undefined;
-            model?: string | undefined;
-            fillOption?: {
-                style: "aspectFit" | "aspectFill";
-            } | undefined;
-            transition?: {
-                type: "fade" | "slideout_left";
-                duration: number;
-            } | undefined;
-        } | undefined;
         lipSyncParams?: {
             provider?: string | undefined;
             model?: string | undefined;
@@ -371,6 +371,17 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                     prompt: string;
                 }> | undefined;
             };
+            movieParams: {
+                provider?: string | undefined;
+                model?: string | undefined;
+                fillOption?: {
+                    style: "aspectFit" | "aspectFill";
+                } | undefined;
+                transition?: {
+                    type: "fade" | "slideout_left";
+                    duration: number;
+                } | undefined;
+            };
             soundEffectParams: {
                 provider?: string | undefined;
                 model?: string | undefined;
@@ -567,17 +578,6 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                 enableLipSync?: boolean | undefined;
                 hidden?: boolean | undefined;
             }[];
-            movieParams?: {
-                provider?: string | undefined;
-                model?: string | undefined;
-                fillOption?: {
-                    style: "aspectFit" | "aspectFill";
-                } | undefined;
-                transition?: {
-                    type: "fade" | "slideout_left";
-                    duration: number;
-                } | undefined;
-            } | undefined;
             lipSyncParams?: {
                 provider?: string | undefined;
                 model?: string | undefined;
@@ -693,6 +693,17 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                 prompt: string;
             }> | undefined;
         };
+        movieParams: {
+            provider?: string | undefined;
+            model?: string | undefined;
+            fillOption?: {
+                style: "aspectFit" | "aspectFill";
+            } | undefined;
+            transition?: {
+                type: "fade" | "slideout_left";
+                duration: number;
+            } | undefined;
+        };
         soundEffectParams: {
             provider?: string | undefined;
             model?: string | undefined;
@@ -716,17 +727,6 @@ export declare const initializeContextFromFiles: (files: FileObject, raiseError:
                 path: string;
             } | undefined;
         };
-        movieParams?: {
-            provider?: string | undefined;
-            model?: string | undefined;
-            fillOption?: {
-                style: "aspectFit" | "aspectFill";
-            } | undefined;
-            transition?: {
-                type: "fade" | "slideout_left";
-                duration: number;
-            } | undefined;
-        } | undefined;
         lipSyncParams?: {
             provider?: string | undefined;
             model?: string | undefined;

package/lib/utils/provider2agent.d.ts

@@ -76,7 +76,10 @@ export declare const provider2MovieAgent: {
         models: string[];
         keyName: string;
         modelParams: {
-            "veo-3.0-generate-preview": {
+            "veo-3.1-generate-preview": {
+                durations: number[];
+            };
+            "veo-3.0-generate-001": {
                 durations: number[];
             };
             "veo-2.0-generate-001": {

package/lib/utils/provider2agent.js

@@ -42,8 +42,8 @@ export const provider2ImageAgent = {
     },
     google: {
         agentName: "imageGenAIAgent",
-        defaultModel: "gemini-2.5-flash-image-preview",
-        models: ["imagen-3.0-generate-002", "imagen-4.0-generate-preview-06-06", "imagen-4.0-ultra-generate-preview-06-06", "gemini-2.5-flash-image-preview"],
+        defaultModel: "gemini-2.5-flash-image",
+        models: ["imagen-4.0-generate-preview-06-06", "imagen-4.0-ultra-generate-preview-06-06", "gemini-2.5-flash-image", "gemini-3-pro-image-preview"],
         keyName: "GEMINI_API_KEY",
     },
     replicate: {
@@ -77,8 +77,8 @@ export const provider2MovieAgent = {
             "minimax/hailuo-02",
             "minimax/hailuo-02-fast",
             "pixverse/pixverse-v4.5",
-            "wan-video/wan-2.2-i2v-480p-fast",
-            "wan-video/wan-2.2-t2v-480p-fast",
+            "wan-video/wan-2.2-i2v-fast",
+            "wan-video/wan-2.2-t2v-fast",
         ],
         modelParams: {
             "bytedance/seedance-1-lite": {
@@ -144,12 +144,12 @@ export const provider2MovieAgent = {
                 last_image: "last_frame_image",
                 price_per_sec: 0.12,
             },
-            "wan-video/wan-2.2-i2v-480p-fast": {
+            "wan-video/wan-2.2-i2v-fast": {
                 durations: [5],
                 start_image: "image",
                 price_per_sec: 0.012,
             },
-            "wan-video/wan-2.2-t2v-480p-fast": {
+            "wan-video/wan-2.2-t2v-fast": {
                 durations: [5],
                 start_image: undefined,
                 price_per_sec: 0.012,
@@ -159,10 +159,13 @@ export const provider2MovieAgent = {
     google: {
         agentName: "movieGenAIAgent",
         defaultModel: "veo-2.0-generate-001",
-        models: ["veo-2.0-generate-001", "veo-3.0-generate-preview"],
+        models: ["veo-2.0-generate-001", "veo-3.0-generate-001", "veo-3.1-generate-preview"],
         keyName: "GEMINI_API_KEY",
         modelParams: {
-            "veo-3.0-generate-preview": {
+            "veo-3.1-generate-preview": {
+                durations: [4, 6, 8],
+            },
+            "veo-3.0-generate-001": {
                 durations: [4, 6, 8],
             },
             "veo-2.0-generate-001": {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mulmocast",
-  "version": "2.0.2",
+  "version": "2.0.3",
   "description": "",
   "type": "module",
   "main": "lib/index.node.js",
@@ -74,7 +74,7 @@
   "homepage": "https://github.com/receptron/mulmocast-cli#readme",
   "dependencies": {
     "@google-cloud/text-to-speech": "^6.4.0",
-    "@google/genai": "^1.29.1",
+    "@google/genai": "^1.30.0",
     "@graphai/anthropic_agent": "^2.0.11",
     "@graphai/browserless_agent": "^2.0.1",
     "@graphai/gemini_agent": "^2.0.1",
@@ -95,11 +95,11 @@
     "fluent-ffmpeg": "^2.1.3",
     "graphai": "^2.0.16",
     "jsdom": "^27.2.0",
-    "marked": "^17.0.0",
+    "marked": "^17.0.1",
     "mulmocast-vision": "^1.0.4",
     "ora": "^9.0.0",
-    "puppeteer": "^24.30.0",
-    "replicate": "^1.3.1",
+    "puppeteer": "^24.31.0",
+    "replicate": "^1.4.0",
     "yaml": "^2.8.1",
     "yargs": "^18.0.0",
     "zod": "^4.1.12"
@@ -117,7 +117,7 @@
     "prettier": "^3.6.2",
     "tsx": "^4.20.6",
     "typescript": "^5.9.3",
-    "typescript-eslint": "^8.46.4"
+    "typescript-eslint": "^8.47.0"
   },
   "engines": {
     "node": ">=20.0.0"

package/scripts/test/gpt.json

@@ -21,11 +21,17 @@
     }
   },
   "beats": [
+    {
+      "speaker": "Host",
+      "text": "How are you?",
+      "imagePrompt": "A witch in Harajuku"
+    },
     {
       "speaker": "Host",
       "text": "How are you?",
       "imagePrompt": "A witch in Harajuku",
       "imageParams": {
+        "model": "gpt-image-1-mini",
         "style": "Ukiyoe-style"
       }
     }

package/scripts/test/test_genai.json

@@ -10,19 +10,19 @@
   "lang": "en",
   "beats": [
     {
-      "id": "gemini_2_5_flash_image_preview",
-      "text": "image generated by gemini-2.5-flash-image-preview",
+      "id": "gemini_3_pro_image_preview",
+      "text": "image generated by gemini-3-pro-image-preview",
       "imagePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
       "imageParams": {
-        "model": "gemini-2.5-flash-image-preview"
+        "model": "gemini-3-pro-image-preview"
       }
     },
     {
-      "id": "imagen_3",
-      "text": "image generated by imagen-3",
+      "id": "gemini_2_5_flash_image",
+      "text": "image generated by gemini-2.5-flash-image",
       "imagePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
       "imageParams": {
-        "model": "imagen-3.0-generate-002"
+        "model": "gemini-2.5-flash-image"
       }
     },
     {
@@ -56,7 +56,15 @@
       "text": "movie generated by veo3",
       "moviePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
       "movieParams": {
-        "model": "veo-3.0-generate-preview"
+        "model": "veo-3.0-generate-001"
+      }
+    },
+    {
+      "id": "genai_veo3_1",
+      "text": "movie generated by veo3_1",
+      "moviePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
+      "movieParams": {
+        "model": "veo-3.1-generate-preview"
       }
     },
     {
@@ -65,11 +73,11 @@
       "duration": 5,
       "imagePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
       "imageParams": {
-        "model": "gemini-2.5-flash-image-preview"
+        "model": "gemini-2.5-flash-image"
       },
       "moviePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
       "movieParams": {
-        "model": "veo-3.0-generate-preview"
+        "model": "veo-3.0-generate-001"
       }
     }
   ]

package/scripts/test/test_image_refs.json

@@ -53,7 +53,7 @@
       "imageNames": ["witch", "broom"],
       "imageParams": {
         "provider": "google",
-        "model": "gemini-2.5-flash-image-preview"
+        "model": "gemini-2.5-flash-image"
       }
     },
     {

package/scripts/test/test_lipsync2.json

@@ -0,0 +1,66 @@
+{
+  "$mulmocast": {
+    "version": "1.1",
+    "credit": "closing"
+  },
+  "canvasSize": {
+    "width": 1280,
+    "height": 720
+  },
+  "speechParams": {
+    "speakers": {
+      "Presenter": {
+        "displayName": {
+          "en": "Presenter"
+        },
+        "voiceId": "shimmer",
+        "provider": "openai"
+      }
+    }
+  },
+  "imageParams": {
+    "provider": "openai",
+    "images": {
+      "woman": {
+        "type": "image",
+        "source": {
+          "kind": "path",
+          "path": "./image-2.png"
+        }
+      }
+    }
+  },
+  "movieParams": {
+    "provider": "replicate",
+    "model": "bytedance/seedance-1-lite"
+  },
+  "soundEffectParams": {
+    "provider": "replicate"
+  },
+  "lipSyncParams": {
+    "provider": "replicate",
+    "model": "bytedance/latentsync"
+  },
+  "audioParams": {
+    "padding": 0.3,
+    "introPadding": 1,
+    "closingPadding": 0.8,
+    "outroPadding": 1,
+    "bgmVolume": 0.2,
+    "audioVolume": 1,
+    "suppressSpeech": false
+  },
+  "title": "リップシンク（generate contents）",
+  "description": "mulmocast",
+  "lang": "en",
+  "beats": [
+    {
+      "speaker": "Presenter",
+      "text": "You've had a challenging day. The evidence is clear: you need rest. You're more capable than you realize. Case closed. hello hello !!",
+      "id": "c0e344e1-4ed7-4145-b7fc-c878c093cffd",
+      "imagePrompt": "in the city",
+      "moviePrompt": "never looking away from camera",
+      "enableLipSync": true
+    }
+  ]
+}

package/scripts/test/test_lipsync2.json~

@@ -0,0 +1,24 @@
+{
+  "$mulmocast": {
+    "version": "1.1"
+  },
+  "lang": "en",
+  "movieParams": {
+    "provider": "google"
+  },
+  "lipSyncParams": {
+    "provider": "replicate",
+    "model": "bytedance/latentsync"
+  },
+  "canvasSize": {
+    "width": 1280,
+    "height": 720
+  },
+  "beats": [
+    {
+      "text": "Hello, I'm macoro. Hello, I'm macoro. Hello, I'm macoro.",
+      "moviePrompt": "macoro is speaking",
+      "enableLipSync": true
+    }
+  ]
+}

package/scripts/test/test_replicate.json

@@ -143,22 +143,22 @@
       }
     },
     {
-      "id": "wan-2.2-i2v-480p-fast",
-      "text": "wan-video/wan-2.2-i2v-480p-fast",
+      "id": "wan-2.2-i2v-fast",
+      "text": "wan-video/wan-2.2-i2v-fast",
       "duration": 5,
       "imagePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
       "moviePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
       "movieParams": {
-        "model": "wan-video/wan-2.2-i2v-480p-fast"
+        "model": "wan-video/wan-2.2-i2v-fast"
       }
     },
     {
-      "id": "wan-2.2-t2v-480p-fast",
-      "text": "wan-video/wan-2.2-t2v-480p-fast",
+      "id": "wan-2.2-t2v-fast",
+      "text": "wan-video/wan-2.2-t2v-fast",
       "duration": 5,
       "moviePrompt": "a woman is walking through a busy Tokyo street at night, she is wearing dark sunglasses",
       "movieParams": {
-        "model": "wan-video/wan-2.2-t2v-480p-fast"
+        "model": "wan-video/wan-2.2-t2v-fast"
       }
     },
     {