mulmocast 0.0.9 → 0.0.10

package/assets/templates/akira_comic.json

@@ -1,6 +1,6 @@
 {
-  "title": "Dr. Slump Style Comic Strips",
-  "description": "Template for Dr. Slump-style comic.",
+  "title": "Akira style",
+  "description": "Template for Akira style comic presentation.",
   "systemPrompt": "Generate a script for a presentation of the given topic. 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.",
   "presentationStyle": {
     "$mulmocast": {

package/assets/templates/drslump_comic.json

@@ -1,6 +1,6 @@
 {
-  "title": "Dr. Slump Style Comic Strips",
-  "description": "Template for Dr. Slump-style comic.",
+  "title": "Dr. Slump Style",
+  "description": "Template for Dr. Slump style comic presentation.",
   "systemPrompt": "Generate a script for a presentation of the given topic. 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.",
   "presentationStyle": {
     "$mulmocast": {

package/assets/templates/ghibli_comic.json

@@ -1,6 +1,6 @@
 {
-  "title": "American Comic Strips",
-  "description": "Template for Dilbert-style comic strips.",
+  "title": "Ghibli comic style",
+  "description": "Template for Ghibli-style comic presentation.",
   "systemPrompt": "Generate a script for a presentation of the given topic. 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.",
   "presentationStyle": {
     "$mulmocast": {

package/assets/templates/ghost_comic.json

@@ -1,6 +1,6 @@
 {
-  "title": "Dr. Slump Style Comic Strips",
-  "description": "Template for Dr. Slump-style comic.",
+  "title": "Ghost in the shell style",
+  "description": "Template for Ghost in the shell style comic presentation.",
   "systemPrompt": "Generate a script for a presentation of the given topic. 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.",
   "presentationStyle": {
     "$mulmocast": {

package/assets/templates/onepiece_comic.json

@@ -1,6 +1,6 @@
 {
-  "title": "Dr. Slump Style Comic Strips",
-  "description": "Template for Dr. Slump-style comic.",
+  "title": "One Piece style",
+  "description": "Template for One Piece style comic presentation.",
   "systemPrompt": "Generate a script for a presentation of the given topic. 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.",
   "presentationStyle": {
     "$mulmocast": {

package/assets/templates/portrait_movie.json

@@ -0,0 +1,28 @@
+{
+  "title": "Photo realistic movie (portrait)",
+  "description": "Template for photo realistic movie in portrait mode.",
+  "systemPrompt": "Generate a script for a presentation of the given topic. Another AI will generate images for each beat based on the image prompt of that beat. Movie prompts must be written in English. Mention the reference in one of beats, if it exists. Use the JSON below as a template.",
+  "presentationStyle": {
+    "$mulmocast": {
+      "version": "1.0",
+      "credit": "closing"
+    },
+    "canvasSize": {
+      "width": 1024,
+      "height": 1536
+    },
+    "imageParams": {
+      "style": "<style>Photo realistic, cinematic.</style>",
+      "images": {
+        "presenter": {
+          "type": "image",
+          "source": {
+            "kind": "url",
+            "url": "https://raw.githubusercontent.com/receptron/mulmocast-media/refs/heads/main/characters/female_presenter.png"
+          }
+        }
+      }
+    }
+  },
+  "scriptName": "movie_prompts_template.json"
+}

package/assets/templates/realistic_movie.json

@@ -0,0 +1,28 @@
+{
+  "title": "Photo realistic movie template",
+  "description": "Template for photo realistic movie.",
+  "systemPrompt": "Generate a script for a presentation of the given topic. Another AI will generate images for each beat based on the image prompt of that beat. Movie prompts must be written in English. Mention the reference in one of beats, if it exists. Use the JSON below as a template.",
+  "presentationStyle": {
+    "$mulmocast": {
+      "version": "1.0",
+      "credit": "closing"
+    },
+    "canvasSize": {
+      "width": 1536,
+      "height": 1024
+    },
+    "imageParams": {
+      "style": "<style>Photo realistic, cinematic.</style>",
+      "images": {
+        "presenter": {
+          "type": "image",
+          "source": {
+            "kind": "url",
+            "url": "https://raw.githubusercontent.com/receptron/mulmocast-media/refs/heads/main/characters/female_presenter.png"
+          }
+        }
+      }
+    }
+  },
+  "scriptName": "movie_prompts_template.json"
+}

package/lib/actions/audio.js

@@ -1,6 +1,6 @@
 import "dotenv/config";
 import { GraphAI } from "graphai";
-import vanillaAgents from "@graphai/vanilla";
+import * as agents from "@graphai/vanilla";
 import ttsNijivoiceAgent from "../agents/tts_nijivoice_agent.js";
 import addBGMAgent from "../agents/add_bgm_agent.js";
 import combineAudioFilesAgent from "../agents/combine_audio_files_agent.js";
@@ -12,7 +12,7 @@ import { fileCacheAgentFilter } from "../utils/filters.js";
 import { getAudioArtifactFilePath, getAudioSegmentDirPath, getAudioCombinedFilePath, getOutputStudioFilePath, defaultBGMPath, mkdir, writingMessage, getAudioSegmentFilePath, resolveMediaSource, } from "../utils/file.js";
 import { text2hash, localizedText } from "../utils/utils.js";
 import { MulmoStudioMethods } from "../methods/mulmo_studio.js";
-// const { default: __, ...vanillaAgents } = agents;
+const vanillaAgents = agents.default ?? agents;
 // const rion_takanashi_voice = "b9277ce3-ba1c-4f6f-9a65-c05ca102ded0"; // たかなし りおん
 // const ben_carter_voice = "bc06c63f-fef6-43b6-92f7-67f919bd5dae"; // ベン・カーター
 const provider_to_agent = {

package/lib/actions/captions.js

@@ -1,9 +1,9 @@
 import { GraphAI, GraphAILogger } from "graphai";
-import vanillaAgents from "@graphai/vanilla";
+import * as agents from "@graphai/vanilla";
 import { getHTMLFile } from "../utils/file.js";
 import { renderHTMLToImage, interpolate } from "../utils/markdown.js";
 import { MulmoStudioMethods } from "../methods/mulmo_studio.js";
-// const { default: __, ...vanillaAgents } = agents;
+const vanillaAgents = agents.default ?? agents;
 const graph_data = {
     version: 0.5,
     nodes: {

package/lib/actions/images.js

@@ -1,16 +1,17 @@
 import dotenv from "dotenv";
 import fs from "fs";
 import { GraphAI, GraphAILogger } from "graphai";
-import vanillaAgents from "@graphai/vanilla";
+import * as agents from "@graphai/vanilla";
 import { fileWriteAgent } from "@graphai/vanilla_node_agents";
 import { getOutputStudioFilePath, mkdir } from "../utils/file.js";
 import { fileCacheAgentFilter } from "../utils/filters.js";
 import imageGoogleAgent from "../agents/image_google_agent.js";
 import imageOpenaiAgent from "../agents/image_openai_agent.js";
+import movieGoogleAgent from "../agents/movie_google_agent.js";
 import { MulmoScriptMethods, MulmoStudioContextMethods } from "../methods/index.js";
 import { imagePlugins } from "../utils/image_plugins/index.js";
 import { imagePrompt } from "../utils/prompt.js";
-// const { default: __, ...vanillaAgents } = agents;
+const vanillaAgents = agents.default ?? agents;
 dotenv.config();
 // const openai = new OpenAI();
 import { GoogleAuth } from "google-auth-library";
@@ -24,6 +25,10 @@ const htmlStyle = (script, beat) => {
 const imagePreprocessAgent = async (namedInputs) => {
     const { context, beat, index, suffix, imageDirPath, imageAgentInfo, imageRefs } = namedInputs;
     const imageParams = { ...imageAgentInfo.imageParams, ...beat.imageParams };
+    if (!imageParams.size) {
+        const canvasSize = MulmoScriptMethods.getCanvasSize(context.studio.script);
+        imageParams.size = `${canvasSize.width}x${canvasSize.height}`;
+    }
     const imagePath = `${imageDirPath}/${context.studio.filename}/${index}${suffix}.png`;
     const returnValue = {
         aspectRatio: MulmoScriptMethods.getAspectRatio(context.studio.script),
@@ -111,14 +116,48 @@ const graph_data = {
                         },
                         defaultValue: {},
                     },
+                    prepareMovie: {
+                        agent: (namedInputs) => {
+                            const { beat, imageDirPath, index, context } = namedInputs;
+                            if (beat.moviePrompt) {
+                                const movieFile = `${imageDirPath}/${context.studio.filename}/${index}.mov`;
+                                return { movieFile };
+                            }
+                            return {};
+                        },
+                        inputs: {
+                            result: ":imageGenerator", // to wait for imageGenerator to finish
+                            imagePath: ":preprocessor.path",
+                            beat: ":beat",
+                            imageDirPath: ":imageDirPath",
+                            index: ":__mapIndex",
+                            context: ":context",
+                        },
+                    },
+                    movieGenerator: {
+                        if: ":prepareMovie.movieFile",
+                        agent: "movieGoogleAgent",
+                        inputs: {
+                            prompt: ":beat.moviePrompt",
+                            imagePath: ":preprocessor.path",
+                            file: ":prepareMovie.movieFile",
+                            studio: ":context.studio", // for cache
+                            index: ":__mapIndex", // for cache
+                            sessionType: "movie", // for cache
+                            params: {
+                                model: ":context.studio.script.movieParams.model",
+                                aspectRatio: ":preprocessor.aspectRatio",
+                                duration: ":beat.duration",
+                            },
+                        },
+                        defaultValue: {},
+                    },
                     output: {
                         agent: "copyAgent",
                         inputs: {
-                            result: ":imageGenerator",
-                            image: ":preprocessor.path",
-                        },
-                        output: {
-                            imageFile: ".image",
+                            onComplete: ":movieGenerator",
+                            imageFile: ":preprocessor.path",
+                            movieFile: ":prepareMovie.movieFile",
                         },
                         isResult: true,
                     },
@@ -167,7 +206,7 @@ const generateImages = async (context) => {
         {
             name: "fileCacheAgentFilter",
             agent: fileCacheAgentFilter,
-            nodeIds: ["imageGenerator"],
+            nodeIds: ["imageGenerator", "movieGenerator"],
         },
     ];
     const options = {
@@ -175,7 +214,7 @@ const generateImages = async (context) => {
     };
     const imageAgentInfo = MulmoScriptMethods.getImageAgentInfo(studio.script);
     // We need to get google's auth token only if the google is the text2image provider.
-    if (imageAgentInfo.provider === "google") {
+    if (imageAgentInfo.provider === "google" || studio.script.movieParams?.provider === "google") {
         GraphAILogger.log("google was specified as text2image engine");
         const token = await googleAuth();
         options.config = {
@@ -183,6 +222,10 @@ const generateImages = async (context) => {
                 projectId: process.env.GOOGLE_PROJECT_ID,
                 token,
             },
+            movieGoogleAgent: {
+                projectId: process.env.GOOGLE_PROJECT_ID,
+                token,
+            },
         };
     }
     if (imageAgentInfo.provider === "openai") {
@@ -219,7 +262,7 @@ const generateImages = async (context) => {
         imageDirPath,
         imageRefs,
     };
-    const graph = new GraphAI(graph_data, { ...vanillaAgents, imageGoogleAgent, imageOpenaiAgent, fileWriteAgent }, options);
+    const graph = new GraphAI(graph_data, { ...vanillaAgents, imageGoogleAgent, movieGoogleAgent, imageOpenaiAgent, fileWriteAgent }, options);
     Object.keys(injections).forEach((key) => {
         graph.injectValue(key, injections[key]);
     });

package/lib/actions/movie.js

@@ -74,8 +74,8 @@ const createVideo = async (audioArtifactFilePath, outputVideoPath, studio, capti
         if (!studioBeat.imageFile || !studioBeat.duration) {
             throw new Error(`studioBeat.imageFile or studioBeat.duration is not set: index=${index}`);
         }
-        const inputIndex = FfmpegContextAddInput(ffmpegContext, studioBeat.imageFile);
-        const mediaType = MulmoScriptMethods.getImageType(studio.script, beat);
+        const inputIndex = FfmpegContextAddInput(ffmpegContext, studioBeat.movieFile ?? studioBeat.imageFile);
+        const mediaType = studioBeat.movieFile ? "movie" : MulmoScriptMethods.getImageType(studio.script, beat);
         const extraPadding = (() => {
             // We need to consider only intro and outro padding because the other paddings were already added to the beat.duration
             if (index === 0) {

package/lib/actions/translate.js

@@ -1,13 +1,13 @@
 import "dotenv/config";
 import { GraphAI, assert } from "graphai";
-import vanillaAgents from "@graphai/vanilla";
+import * as agents from "@graphai/vanilla";
 import { openAIAgent } from "@graphai/openai_agent";
 import { fileWriteAgent } from "@graphai/vanilla_node_agents";
 import { recursiveSplitJa, replacementsJa, replacePairsJa } from "../utils/string.js";
 import { getOutputStudioFilePath, mkdir, writingMessage } from "../utils/file.js";
 import { translateSystemPrompt, translatePrompts } from "../utils/prompt.js";
 import { MulmoStudioMethods } from "../methods/mulmo_studio.js";
-// const { default: __, ...vanillaAgents } = agents;
+const vanillaAgents = agents.default ?? agents;
 const translateGraph = {
     version: 0.5,
     nodes: {

package/lib/agents/image_openai_agent.js

@@ -9,7 +9,7 @@ export const imageOpenaiAgent = async ({ namedInputs, params }) => {
         model: model ?? "dall-e-3",
         prompt,
         n: 1,
-        size: size || model === "gpt-image-1" ? "1536x1024" : "1792x1024",
+        size: size ?? (model === "gpt-image-1" ? "1536x1024" : "1792x1024"),
     };
     if (model === "gpt-image-1") {
         imageOptions.moderation = moderation || "auto";

package/lib/agents/movie_google_agent.d.ts

@@ -0,0 +1,17 @@
+import type { AgentFunction, AgentFunctionInfo } from "graphai";
+export type MovieGoogleConfig = {
+    projectId?: string;
+    token?: string;
+};
+export declare const movieGoogleAgent: AgentFunction<{
+    model: string;
+    aspectRatio: string;
+    duration?: number;
+}, {
+    buffer: Buffer;
+}, {
+    prompt: string;
+    imagePath: string;
+}, MovieGoogleConfig>;
+declare const movieGoogleAgentInfo: AgentFunctionInfo;
+export default movieGoogleAgentInfo;

package/lib/agents/movie_google_agent.js

@@ -0,0 +1,114 @@
+import { readFileSync } from "fs";
+import { GraphAILogger, sleep } from "graphai";
+async function generateMovie(projectId, model, token, prompt, imagePath, aspectRatio, duration) {
+    const GOOGLE_IMAGEN_ENDPOINT = `https://us-central1-aiplatform.googleapis.com/v1/projects/${projectId}/locations/us-central1/publishers/google/models/${model}`;
+    // Prepare the payload for the API request
+    const buffer = readFileSync(imagePath);
+    const bytesBase64Encoded = buffer.toString("base64");
+    const payload = {
+        instances: [
+            {
+                prompt: prompt,
+                image: {
+                    bytesBase64Encoded,
+                    mimeType: "image/png",
+                },
+            },
+        ],
+        parameters: {
+            sampleCount: 1,
+            aspectRatio: aspectRatio,
+            //safetySetting: "block_only_high",
+            durationSeconds: duration,
+        },
+    };
+    // Make the API call using fetch
+    const response = await fetch(`${GOOGLE_IMAGEN_ENDPOINT}:predictLongRunning`, {
+        method: "POST",
+        headers: {
+            Authorization: `Bearer ${token}`,
+            "Content-Type": "application/json",
+        },
+        body: JSON.stringify(payload),
+    });
+    if (!response.ok) {
+        throw new Error(`Error: ${response.status} - ${response.statusText}`);
+    }
+    const initialResponse = await response.json();
+    const fetchBody = {
+        operationName: initialResponse.name,
+    };
+    const completeResponse = await (async () => {
+        while (true) {
+            GraphAILogger.info("...waiting for movie generation...");
+            await sleep(3000);
+            const response = await fetch(`${GOOGLE_IMAGEN_ENDPOINT}:fetchPredictOperation`, {
+                method: "POST",
+                headers: {
+                    Authorization: `Bearer ${token}`,
+                    "Content-Type": "application/json",
+                },
+                body: JSON.stringify(fetchBody),
+            });
+            if (!response.ok) {
+                throw new Error(`Error: ${response.status} - ${response.statusText}`);
+            }
+            const responseData = await response.json();
+            if (responseData.done) {
+                if (responseData.error) {
+                    GraphAILogger.info("Prompt: ", prompt);
+                    throw new Error(`Error: ${responseData.error.message}`);
+                }
+                if (!responseData.response.videos) {
+                    throw new Error(`No video: ${JSON.stringify(responseData, null, 2)}`);
+                }
+                return responseData.response;
+            }
+        }
+    })();
+    const encodedMovie = completeResponse.videos[0].bytesBase64Encoded;
+    if (encodedMovie) {
+        return Buffer.from(encodedMovie, "base64");
+    }
+    return undefined;
+}
+export const movieGoogleAgent = async ({ namedInputs, params, config }) => {
+    const { prompt, imagePath } = namedInputs;
+    /*
+    if (prompt) {
+      const buffer = Buffer.from(prompt);
+      return { buffer };
+    }
+    */
+    const aspectRatio = params.aspectRatio ?? "16:9";
+    const model = params.model ?? "veo-2.0-generate-001"; // "veo-3.0-generate-preview";
+    const duration = params.duration ?? 8;
+    //const projectId = process.env.GOOGLE_PROJECT_ID; // Your Google Cloud Project ID
+    const projectId = config?.projectId;
+    const token = config?.token;
+    try {
+        const buffer = await generateMovie(projectId, model, token, prompt, imagePath, aspectRatio, duration);
+        if (buffer) {
+            return { buffer };
+        }
+        throw new Error("ERROR: geneateImage returned undefined");
+    }
+    catch (error) {
+        GraphAILogger.info("Failed to generate movie:", error);
+        throw error;
+    }
+};
+const movieGoogleAgentInfo = {
+    name: "movieGoogleAgent",
+    agent: movieGoogleAgent,
+    mock: movieGoogleAgent,
+    samples: [],
+    description: "Google Movie agent",
+    category: ["movie"],
+    author: "Receptron Team",
+    repository: "https://github.com/receptron/mulmocast-cli/",
+    // source: "https://github.com/receptron/mulmocast-cli/blob/main/src/agents/image_google_agent.ts",
+    license: "MIT",
+    environmentVariables: [],
+};
+export default movieGoogleAgentInfo;

package/lib/cli/bin.js

@@ -36,6 +36,18 @@ export const main = async () => {
         .demandCommand()
         .strict()
         .help()
+        .showHelpOnFail(false)
+        .fail((msg, err, y) => {
+        // if yargs detect error, show help and exit
+        if (msg) {
+            y.showHelp();
+            GraphAILogger.info("\\n" + msg);
+            process.exit(1);
+        }
+        if (err) {
+            throw err;
+        }
+    })
         .alias("help", "h");
     await cli.parseAsync();
 };

package/lib/methods/mulmo_studio.d.ts

@@ -1,6 +1,6 @@
 import { MulmoStudio } from "../types/index.js";
 type SessionType = "audio" | "image" | "video" | "multiLingual" | "caption" | "pdf";
-type BeatSessionType = "audio" | "image" | "multiLingual" | "caption";
+type BeatSessionType = "audio" | "image" | "multiLingual" | "caption" | "movie";
 export declare const MulmoStudioMethods: {
     setSessionState(studio: MulmoStudio, sessionType: SessionType, value: boolean): void;
     setBeatSessionState(studio: MulmoStudio, sessionType: BeatSessionType, index: number, value: boolean): void;

package/lib/tools/create_mulmo_script_from_url.js

@@ -4,7 +4,7 @@ import { openAIAgent } from "@graphai/openai_agent";
 import { anthropicAgent } from "@graphai/anthropic_agent";
 import { geminiAgent } from "@graphai/gemini_agent";
 import { groqAgent } from "@graphai/groq_agent";
-import vanillaAgents from "@graphai/vanilla";
+import * as agents from "@graphai/vanilla";
 import { fileWriteAgent } from "@graphai/vanilla_node_agents";
 import { browserlessAgent } from "@graphai/browserless_agent";
 import validateSchemaAgent from "../agents/validate_schema_agent.js";
@@ -14,7 +14,7 @@ import { mulmoScriptSchema, urlsSchema } from "../types/schema.js";
 import { cliLoadingPlugin } from "../utils/plugins.js";
 import { graphDataScriptFromUrlPrompt } from "../utils/prompt.js";
 import { llmPair } from "../utils/utils.js";
-// const { default: __, ...vanillaAgents } = agents;
+const vanillaAgents = agents.default ?? agents;
 const graphData = {
     version: 0.5,
     // Execute sequentially because the free version of browserless API doesn't support concurrent execution.

package/lib/tools/create_mulmo_script_interactively.js

@@ -6,7 +6,7 @@ import { openAIAgent } from "@graphai/openai_agent";
 import { anthropicAgent } from "@graphai/anthropic_agent";
 import { geminiAgent } from "@graphai/gemini_agent";
 import { groqAgent } from "@graphai/groq_agent";
-import vanillaAgents from "@graphai/vanilla";
+import * as agents from "@graphai/vanilla";
 import { fileWriteAgent } from "@graphai/vanilla_node_agents";
 import { readTemplatePrompt, mkdir } from "../utils/file.js";
 import { browserlessCacheGenerator } from "../utils/filters.js";
@@ -16,7 +16,7 @@ import validateSchemaAgent from "../agents/validate_schema_agent.js";
 import { llmPair } from "../utils/utils.js";
 import { interactiveClarificationPrompt, prefixPrompt } from "../utils/prompt.js";
 // import { cliLoadingPlugin } from "../utils/plugins.js";
-// const { default: __, ...vanillaAgents } = agents;
+const vanillaAgents = agents.default ?? agents;
 const agentHeader = "\x1b[34m● \x1b[0m\x1b[1mAgent\x1b[0m:\x1b[0m";
 const graphDataForScraping = {
     version: 0.5,

package/lib/tools/story_to_script.js

@@ -5,14 +5,14 @@ import { openAIAgent } from "@graphai/openai_agent";
 import { anthropicAgent } from "@graphai/anthropic_agent";
 import { geminiAgent } from "@graphai/gemini_agent";
 import { groqAgent } from "@graphai/groq_agent";
-import vanillaAgents from "@graphai/vanilla";
+import * as agents from "@graphai/vanilla";
 import { graphDataScriptGeneratePrompt, sceneToBeatsPrompt, storyToScriptInfoPrompt, storyToScriptPrompt } from "../utils/prompt.js";
 import { fileWriteAgent } from "@graphai/vanilla_node_agents";
 import validateSchemaAgent from "../agents/validate_schema_agent.js";
 import { llmPair } from "../utils/utils.js";
 import { storyToScriptGenerateMode } from "../utils/const.js";
 import { cliLoadingPlugin } from "../utils/plugins.js";
-// const { default: __, ...vanillaAgents } = agents;
+const vanillaAgents = agents.default ?? agents;
 const createValidatedScriptGraphData = ({ systemPrompt, prompt, schema, llmAgent, llmModel, maxTokens, }) => {
     return {
         loop: {

package/lib/types/schema.d.ts

@@ -1624,6 +1624,7 @@ export declare const mulmoBeatSchema: z.ZodObject<{
     }>>;
     imageNames: z.ZodOptional<z.ZodArray<z.ZodString, "many">>;
     imagePrompt: z.ZodOptional<z.ZodString>;
+    moviePrompt: z.ZodOptional<z.ZodString>;
 }, "strict", z.ZodTypeAny, {
     text: string;
     speaker: string;
@@ -1780,6 +1781,7 @@ export declare const mulmoBeatSchema: z.ZodObject<{
     } | undefined;
     imageNames?: string[] | undefined;
     imagePrompt?: string | undefined;
+    moviePrompt?: string | undefined;
 }, {
     text: string;
     duration?: number | undefined;
@@ -1936,6 +1938,7 @@ export declare const mulmoBeatSchema: z.ZodObject<{
     } | undefined;
     imageNames?: string[] | undefined;
     imagePrompt?: string | undefined;
+    moviePrompt?: string | undefined;
 }>;
 export declare const mulmoCanvasDimensionSchema: z.ZodDefault<z.ZodObject<{
     width: z.ZodNumber;
@@ -2010,6 +2013,17 @@ export declare const mulmoSpeechParamsSchema: z.ZodObject<{
     provider?: "openai" | "nijivoice" | "google" | undefined;
 }>;
 export declare const text2ImageProviderSchema: z.ZodDefault<z.ZodUnion<[z.ZodLiteral<"openai">, z.ZodLiteral<"google">]>>;
+export declare const text2MovieProviderSchema: z.ZodDefault<z.ZodUnion<[z.ZodLiteral<"openai">, z.ZodLiteral<"google">]>>;
+export declare const mulmoMovieParamsSchema: z.ZodObject<{
+    provider: z.ZodOptional<z.ZodDefault<z.ZodUnion<[z.ZodLiteral<"openai">, z.ZodLiteral<"google">]>>>;
+    model: z.ZodOptional<z.ZodString>;
+}, "strict", z.ZodTypeAny, {
+    model?: string | undefined;
+    provider?: "openai" | "google" | undefined;
+}, {
+    model?: string | undefined;
+    provider?: "openai" | "google" | undefined;
+}>;
 export declare const mulmoPresentationStyleSchema: z.ZodObject<{
     $mulmocast: z.ZodObject<{
         version: z.ZodLiteral<"1.0">;
@@ -2204,6 +2218,16 @@ export declare const mulmoPresentationStyleSchema: z.ZodObject<{
         }> | undefined;
         provider?: "openai" | "google" | undefined;
     }>>;
+    movieParams: z.ZodOptional<z.ZodObject<{
+        provider: z.ZodOptional<z.ZodDefault<z.ZodUnion<[z.ZodLiteral<"openai">, z.ZodLiteral<"google">]>>>;
+        model: z.ZodOptional<z.ZodString>;
+    }, "strict", z.ZodTypeAny, {
+        model?: string | undefined;
+        provider?: "openai" | "google" | undefined;
+    }, {
+        model?: string | undefined;
+        provider?: "openai" | "google" | undefined;
+    }>>;
     textSlideParams: z.ZodOptional<z.ZodObject<{
         cssStyles: z.ZodUnion<[z.ZodString, z.ZodArray<z.ZodString, "many">]>;
     }, "strict", z.ZodTypeAny, {
@@ -2280,6 +2304,10 @@ export declare const mulmoPresentationStyleSchema: z.ZodObject<{
     textSlideParams?: {
         cssStyles: string | string[];
     } | undefined;
+    movieParams?: {
+        model?: string | undefined;
+        provider?: "openai" | "google" | undefined;
+    } | undefined;
     omitCaptions?: boolean | undefined;
 }, {
     $mulmocast: {
@@ -2333,6 +2361,10 @@ export declare const mulmoPresentationStyleSchema: z.ZodObject<{
         }>;
         provider?: "openai" | "nijivoice" | "google" | undefined;
     } | undefined;
+    movieParams?: {
+        model?: string | undefined;
+        provider?: "openai" | "google" | undefined;
+    } | undefined;
     omitCaptions?: boolean | undefined;
 }>;
 export declare const mulmoReferenceSchema: z.ZodObject<{
@@ -2545,6 +2577,16 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         }> | undefined;
         provider?: "openai" | "google" | undefined;
     }>>;
+    movieParams: z.ZodOptional<z.ZodObject<{
+        provider: z.ZodOptional<z.ZodDefault<z.ZodUnion<[z.ZodLiteral<"openai">, z.ZodLiteral<"google">]>>>;
+        model: z.ZodOptional<z.ZodString>;
+    }, "strict", z.ZodTypeAny, {
+        model?: string | undefined;
+        provider?: "openai" | "google" | undefined;
+    }, {
+        model?: string | undefined;
+        provider?: "openai" | "google" | undefined;
+    }>>;
     textSlideParams: z.ZodOptional<z.ZodObject<{
         cssStyles: z.ZodUnion<[z.ZodString, z.ZodArray<z.ZodString, "many">]>;
     }, "strict", z.ZodTypeAny, {
@@ -3240,6 +3282,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         }>>;
         imageNames: z.ZodOptional<z.ZodArray<z.ZodString, "many">>;
         imagePrompt: z.ZodOptional<z.ZodString>;
+        moviePrompt: z.ZodOptional<z.ZodString>;
     }, "strict", z.ZodTypeAny, {
         text: string;
         speaker: string;
@@ -3396,6 +3439,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         } | undefined;
         imageNames?: string[] | undefined;
         imagePrompt?: string | undefined;
+        moviePrompt?: string | undefined;
     }, {
         text: string;
         duration?: number | undefined;
@@ -3552,6 +3596,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         } | undefined;
         imageNames?: string[] | undefined;
         imagePrompt?: string | undefined;
+        moviePrompt?: string | undefined;
     }>, "many">;
     imagePath: z.ZodOptional<z.ZodString>;
     __test_invalid__: z.ZodOptional<z.ZodBoolean>;
@@ -3737,6 +3782,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         } | undefined;
         imageNames?: string[] | undefined;
         imagePrompt?: string | undefined;
+        moviePrompt?: string | undefined;
     }[];
     lang?: string | undefined;
     title?: string | undefined;
@@ -3766,6 +3812,10 @@ export declare const mulmoScriptSchema: z.ZodObject<{
     textSlideParams?: {
         cssStyles: string | string[];
     } | undefined;
+    movieParams?: {
+        model?: string | undefined;
+        provider?: "openai" | "google" | undefined;
+    } | undefined;
     omitCaptions?: boolean | undefined;
     description?: string | undefined;
     references?: {
@@ -3937,6 +3987,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         } | undefined;
         imageNames?: string[] | undefined;
         imagePrompt?: string | undefined;
+        moviePrompt?: string | undefined;
     }[];
     lang?: string | undefined;
     title?: string | undefined;
@@ -3987,6 +4038,10 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         }>;
         provider?: "openai" | "nijivoice" | "google" | undefined;
     } | undefined;
+    movieParams?: {
+        model?: string | undefined;
+        provider?: "openai" | "google" | undefined;
+    } | undefined;
     omitCaptions?: boolean | undefined;
     description?: string | undefined;
     references?: {
@@ -4003,18 +4058,21 @@ export declare const mulmoStudioBeatSchema: z.ZodObject<{
     duration: z.ZodOptional<z.ZodNumber>;
     audioFile: z.ZodOptional<z.ZodString>;
     imageFile: z.ZodOptional<z.ZodString>;
+    movieFile: z.ZodOptional<z.ZodString>;
     captionFile: z.ZodOptional<z.ZodString>;
 }, "strict", z.ZodTypeAny, {
     duration?: number | undefined;
     hash?: string | undefined;
     audioFile?: string | undefined;
     imageFile?: string | undefined;
+    movieFile?: string | undefined;
     captionFile?: string | undefined;
 }, {
     duration?: number | undefined;
     hash?: string | undefined;
     audioFile?: string | undefined;
     imageFile?: string | undefined;
+    movieFile?: string | undefined;
     captionFile?: string | undefined;
 }>;
 export declare const mulmoStudioMultiLingualDataSchema: z.ZodObject<{
@@ -4117,15 +4175,18 @@ export declare const mulmoSessionStateSchema: z.ZodObject<{
     inBeatSession: z.ZodObject<{
         audio: z.ZodSet<z.ZodNumber>;
         image: z.ZodSet<z.ZodNumber>;
+        movie: z.ZodSet<z.ZodNumber>;
         multiLingual: z.ZodSet<z.ZodNumber>;
         caption: z.ZodSet<z.ZodNumber>;
     }, "strip", z.ZodTypeAny, {
         image: Set<number>;
+        movie: Set<number>;
         audio: Set<number>;
         multiLingual: Set<number>;
         caption: Set<number>;
     }, {
         image: Set<number>;
+        movie: Set<number>;
         audio: Set<number>;
         multiLingual: Set<number>;
         caption: Set<number>;
@@ -4141,6 +4202,7 @@ export declare const mulmoSessionStateSchema: z.ZodObject<{
     };
     inBeatSession: {
         image: Set<number>;
+        movie: Set<number>;
         audio: Set<number>;
         multiLingual: Set<number>;
         caption: Set<number>;
@@ -4156,6 +4218,7 @@ export declare const mulmoSessionStateSchema: z.ZodObject<{
     };
     inBeatSession: {
         image: Set<number>;
+        movie: Set<number>;
         audio: Set<number>;
         multiLingual: Set<number>;
         caption: Set<number>;
@@ -4356,6 +4419,16 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             }> | undefined;
             provider?: "openai" | "google" | undefined;
         }>>;
+        movieParams: z.ZodOptional<z.ZodObject<{
+            provider: z.ZodOptional<z.ZodDefault<z.ZodUnion<[z.ZodLiteral<"openai">, z.ZodLiteral<"google">]>>>;
+            model: z.ZodOptional<z.ZodString>;
+        }, "strict", z.ZodTypeAny, {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        }, {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        }>>;
         textSlideParams: z.ZodOptional<z.ZodObject<{
             cssStyles: z.ZodUnion<[z.ZodString, z.ZodArray<z.ZodString, "many">]>;
         }, "strict", z.ZodTypeAny, {
@@ -5051,6 +5124,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             }>>;
             imageNames: z.ZodOptional<z.ZodArray<z.ZodString, "many">>;
             imagePrompt: z.ZodOptional<z.ZodString>;
+            moviePrompt: z.ZodOptional<z.ZodString>;
         }, "strict", z.ZodTypeAny, {
             text: string;
             speaker: string;
@@ -5207,6 +5281,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             imageNames?: string[] | undefined;
             imagePrompt?: string | undefined;
+            moviePrompt?: string | undefined;
         }, {
             text: string;
             duration?: number | undefined;
@@ -5363,6 +5438,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             imageNames?: string[] | undefined;
             imagePrompt?: string | undefined;
+            moviePrompt?: string | undefined;
         }>, "many">;
         imagePath: z.ZodOptional<z.ZodString>;
         __test_invalid__: z.ZodOptional<z.ZodBoolean>;
@@ -5548,6 +5624,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             imageNames?: string[] | undefined;
             imagePrompt?: string | undefined;
+            moviePrompt?: string | undefined;
         }[];
         lang?: string | undefined;
         title?: string | undefined;
@@ -5577,6 +5654,10 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         textSlideParams?: {
             cssStyles: string | string[];
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
         description?: string | undefined;
         references?: {
@@ -5748,6 +5829,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             imageNames?: string[] | undefined;
             imagePrompt?: string | undefined;
+            moviePrompt?: string | undefined;
         }[];
         lang?: string | undefined;
         title?: string | undefined;
@@ -5798,6 +5880,10 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             }>;
             provider?: "openai" | "nijivoice" | "google" | undefined;
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
         description?: string | undefined;
         references?: {
@@ -5815,18 +5901,21 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         duration: z.ZodOptional<z.ZodNumber>;
         audioFile: z.ZodOptional<z.ZodString>;
         imageFile: z.ZodOptional<z.ZodString>;
+        movieFile: z.ZodOptional<z.ZodString>;
         captionFile: z.ZodOptional<z.ZodString>;
     }, "strict", z.ZodTypeAny, {
         duration?: number | undefined;
         hash?: string | undefined;
         audioFile?: string | undefined;
         imageFile?: string | undefined;
+        movieFile?: string | undefined;
         captionFile?: string | undefined;
     }, {
         duration?: number | undefined;
         hash?: string | undefined;
         audioFile?: string | undefined;
         imageFile?: string | undefined;
+        movieFile?: string | undefined;
         captionFile?: string | undefined;
     }>, "many">;
     multiLingual: z.ZodArray<z.ZodObject<{
@@ -5892,15 +5981,18 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         inBeatSession: z.ZodObject<{
             audio: z.ZodSet<z.ZodNumber>;
             image: z.ZodSet<z.ZodNumber>;
+            movie: z.ZodSet<z.ZodNumber>;
             multiLingual: z.ZodSet<z.ZodNumber>;
             caption: z.ZodSet<z.ZodNumber>;
         }, "strip", z.ZodTypeAny, {
             image: Set<number>;
+            movie: Set<number>;
             audio: Set<number>;
             multiLingual: Set<number>;
             caption: Set<number>;
         }, {
             image: Set<number>;
+            movie: Set<number>;
             audio: Set<number>;
             multiLingual: Set<number>;
             caption: Set<number>;
@@ -5916,6 +6008,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         };
         inBeatSession: {
             image: Set<number>;
+            movie: Set<number>;
             audio: Set<number>;
             multiLingual: Set<number>;
             caption: Set<number>;
@@ -5931,6 +6024,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         };
         inBeatSession: {
             image: Set<number>;
+            movie: Set<number>;
             audio: Set<number>;
             multiLingual: Set<number>;
             caption: Set<number>;
@@ -5942,6 +6036,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         hash?: string | undefined;
         audioFile?: string | undefined;
         imageFile?: string | undefined;
+        movieFile?: string | undefined;
         captionFile?: string | undefined;
     }[];
     multiLingual: {
@@ -6135,6 +6230,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             imageNames?: string[] | undefined;
             imagePrompt?: string | undefined;
+            moviePrompt?: string | undefined;
         }[];
         lang?: string | undefined;
         title?: string | undefined;
@@ -6164,6 +6260,10 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         textSlideParams?: {
             cssStyles: string | string[];
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
         description?: string | undefined;
         references?: {
@@ -6187,6 +6287,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         };
         inBeatSession: {
             image: Set<number>;
+            movie: Set<number>;
             audio: Set<number>;
             multiLingual: Set<number>;
             caption: Set<number>;
@@ -6198,6 +6299,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         hash?: string | undefined;
         audioFile?: string | undefined;
         imageFile?: string | undefined;
+        movieFile?: string | undefined;
         captionFile?: string | undefined;
     }[];
     multiLingual: {
@@ -6370,6 +6472,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             imageNames?: string[] | undefined;
             imagePrompt?: string | undefined;
+            moviePrompt?: string | undefined;
         }[];
         lang?: string | undefined;
         title?: string | undefined;
@@ -6420,6 +6523,10 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             }>;
             provider?: "openai" | "nijivoice" | "google" | undefined;
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
         description?: string | undefined;
         references?: {
@@ -6443,6 +6550,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         };
         inBeatSession: {
             image: Set<number>;
+            movie: Set<number>;
             audio: Set<number>;
             multiLingual: Set<number>;
             caption: Set<number>;
@@ -6648,6 +6756,16 @@ export declare const mulmoScriptTemplateSchema: z.ZodObject<{
             }> | undefined;
             provider?: "openai" | "google" | undefined;
         }>>;
+        movieParams: z.ZodOptional<z.ZodObject<{
+            provider: z.ZodOptional<z.ZodDefault<z.ZodUnion<[z.ZodLiteral<"openai">, z.ZodLiteral<"google">]>>>;
+            model: z.ZodOptional<z.ZodString>;
+        }, "strict", z.ZodTypeAny, {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        }, {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        }>>;
         textSlideParams: z.ZodOptional<z.ZodObject<{
             cssStyles: z.ZodUnion<[z.ZodString, z.ZodArray<z.ZodString, "many">]>;
         }, "strict", z.ZodTypeAny, {
@@ -6724,6 +6842,10 @@ export declare const mulmoScriptTemplateSchema: z.ZodObject<{
         textSlideParams?: {
             cssStyles: string | string[];
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
     }, {
         $mulmocast: {
@@ -6777,6 +6899,10 @@ export declare const mulmoScriptTemplateSchema: z.ZodObject<{
             }>;
             provider?: "openai" | "nijivoice" | "google" | undefined;
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
     }>>;
 }, "strict", z.ZodTypeAny, {
@@ -6836,6 +6962,10 @@ export declare const mulmoScriptTemplateSchema: z.ZodObject<{
         textSlideParams?: {
             cssStyles: string | string[];
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
     } | undefined;
 }, {
@@ -6895,6 +7025,10 @@ export declare const mulmoScriptTemplateSchema: z.ZodObject<{
             }>;
             provider?: "openai" | "nijivoice" | "google" | undefined;
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
     } | undefined;
 }>;

package/lib/types/schema.js

@@ -178,7 +178,8 @@ export const mulmoBeatSchema = z
     speechOptions: speechOptionsSchema.optional(),
     textSlideParams: textSlideParamsSchema.optional(),
     imageNames: z.array(imageIdSchema).optional(), // list of image names to use for image generation. The default is all images in the imageParams.images.
-    imagePrompt: z.string().optional(), // specified or inserted by preprocessor
+    imagePrompt: z.string().optional(),
+    moviePrompt: z.string().optional(),
 })
     .strict();
 export const mulmoCanvasDimensionSchema = z
@@ -202,6 +203,13 @@ export const mulmoSpeechParamsSchema = z
 })
     .strict();
 export const text2ImageProviderSchema = z.union([z.literal("openai"), z.literal("google")]).default("openai");
+export const text2MovieProviderSchema = z.union([z.literal("openai"), z.literal("google")]).default("google");
+export const mulmoMovieParamsSchema = z
+    .object({
+    provider: text2MovieProviderSchema.optional(),
+    model: z.string().optional(), // default: provider specific
+})
+    .strict();
 export const mulmoPresentationStyleSchema = z.object({
     $mulmocast: mulmoCastCreditSchema,
     canvasSize: mulmoCanvasDimensionSchema, // has default value
@@ -220,6 +228,7 @@ export const mulmoPresentationStyleSchema = z.object({
         provider: text2ImageProviderSchema, // has default value
     })
         .optional(),
+    movieParams: mulmoMovieParamsSchema.optional(),
     // for textSlides
     textSlideParams: textSlideParamsSchema.optional(),
     // videoParams: videoParamsSchema.optional(),
@@ -257,6 +266,7 @@ export const mulmoStudioBeatSchema = z
     duration: z.number().optional(),
     audioFile: z.string().optional(),
     imageFile: z.string().optional(), // path to the image
+    movieFile: z.string().optional(), // path to the movie file
     captionFile: z.string().optional(), // path to the caption image
 })
     .strict();
@@ -276,6 +286,7 @@ export const mulmoSessionStateSchema = z.object({
     inBeatSession: z.object({
         audio: z.set(z.number()),
         image: z.set(z.number()),
+        movie: z.set(z.number()),
         multiLingual: z.set(z.number()),
         caption: z.set(z.number()),
     }),
@@ -298,6 +309,7 @@ export const mulmoStudioSchema = z
         inBeatSession: {
             audio: new Set(),
             image: new Set(),
+            movie: new Set(),
             multiLingual: new Set(),
             caption: new Set(),
         },

package/lib/types/type.d.ts

@@ -1,4 +1,4 @@
-import { langSchema, localizedTextSchema, mulmoBeatSchema, mulmoScriptSchema, mulmoStudioSchema, mulmoStudioBeatSchema, mulmoStoryboardSchema, mulmoStoryboardSceneSchema, mulmoStudioMultiLingualSchema, mulmoStudioMultiLingualDataSchema, speakerDictionarySchema, mulmoImageParamsSchema, mulmoSpeechParamsSchema, textSlideParamsSchema, speechOptionsSchema, mulmoCanvasDimensionSchema, mulmoScriptTemplateSchema, text2ImageProviderSchema, text2SpeechProviderSchema, mulmoPresentationStyleSchema, multiLingualTextsSchema, mulmoMermaidMediaSchema, mulmoTextSlideMediaSchema, mulmoMarkdownMediaSchema, mulmoImageMediaSchema, mulmoChartMediaSchema, mediaSourceSchema } from "./schema.js";
+import { langSchema, localizedTextSchema, mulmoBeatSchema, mulmoScriptSchema, mulmoStudioSchema, mulmoStudioBeatSchema, mulmoStoryboardSchema, mulmoStoryboardSceneSchema, mulmoStudioMultiLingualSchema, mulmoStudioMultiLingualDataSchema, speakerDictionarySchema, mulmoImageParamsSchema, mulmoMovieParamsSchema, mulmoSpeechParamsSchema, textSlideParamsSchema, speechOptionsSchema, mulmoCanvasDimensionSchema, mulmoScriptTemplateSchema, text2ImageProviderSchema, text2MovieProviderSchema, text2SpeechProviderSchema, mulmoPresentationStyleSchema, multiLingualTextsSchema, mulmoMermaidMediaSchema, mulmoTextSlideMediaSchema, mulmoMarkdownMediaSchema, mulmoImageMediaSchema, mulmoChartMediaSchema, mediaSourceSchema } from "./schema.js";
 import { pdf_modes, pdf_sizes, storyToScriptGenerateMode } from "../utils/const.js";
 import { LLM } from "../utils/utils.js";
 import { z } from "zod";
@@ -10,6 +10,7 @@ export type SpeechOptions = z.infer<typeof speechOptionsSchema>;
 export type MulmoImageParams = z.infer<typeof mulmoImageParamsSchema>;
 export type TextSlideParams = z.infer<typeof textSlideParamsSchema>;
 export type Text2ImageProvider = z.infer<typeof text2ImageProviderSchema>;
+export type Text2MovieProvider = z.infer<typeof text2MovieProviderSchema>;
 export type Text2SpeechProvider = z.infer<typeof text2SpeechProviderSchema>;
 export type LocalizedText = z.infer<typeof localizedTextSchema>;
 export type MulmoScript = z.infer<typeof mulmoScriptSchema>;
@@ -24,6 +25,7 @@ export type MulmoScriptTemplate = z.infer<typeof mulmoScriptTemplateSchema>;
 export type MulmoStudioMultiLingual = z.infer<typeof mulmoStudioMultiLingualSchema>;
 export type MulmoStudioMultiLingualData = z.infer<typeof mulmoStudioMultiLingualDataSchema>;
 export type MultiLingualTexts = z.infer<typeof multiLingualTextsSchema>;
+export type MulmoMovieParams = z.infer<typeof mulmoMovieParamsSchema>;
 export type MulmoTextSlideMedia = z.infer<typeof mulmoTextSlideMediaSchema>;
 export type MulmoMarkdownMedia = z.infer<typeof mulmoMarkdownMediaSchema>;
 export type MulmoImageMedia = z.infer<typeof mulmoImageMediaSchema>;

package/lib/utils/preprocess.d.ts

@@ -5,6 +5,7 @@ export declare const createOrUpdateStudioData: (_mulmoScript: MulmoScript, curre
         hash?: string | undefined;
         audioFile?: string | undefined;
         imageFile?: string | undefined;
+        movieFile?: string | undefined;
         captionFile?: string | undefined;
     }[];
     multiLingual: {
@@ -198,6 +199,7 @@ export declare const createOrUpdateStudioData: (_mulmoScript: MulmoScript, curre
             } | undefined;
             imageNames?: string[] | undefined;
             imagePrompt?: string | undefined;
+            moviePrompt?: string | undefined;
         }[];
         lang?: string | undefined;
         title?: string | undefined;
@@ -227,6 +229,10 @@ export declare const createOrUpdateStudioData: (_mulmoScript: MulmoScript, curre
         textSlideParams?: {
             cssStyles: string | string[];
         } | undefined;
+        movieParams?: {
+            model?: string | undefined;
+            provider?: "openai" | "google" | undefined;
+        } | undefined;
         omitCaptions?: boolean | undefined;
         description?: string | undefined;
         references?: {
@@ -250,6 +256,7 @@ export declare const createOrUpdateStudioData: (_mulmoScript: MulmoScript, curre
         };
         inBeatSession: {
             image: Set<number>;
+            movie: Set<number>;
             audio: Set<number>;
             multiLingual: Set<number>;
             caption: Set<number>;

package/lib/utils/utils.d.ts

@@ -15,3 +15,4 @@ export declare const chunkArray: <T>(array: T[], size?: number) => T[][];
 export declare const isHttp: (fileOrUrl: string) => boolean;
 export declare const text2hash: (input: string) => string;
 export declare const localizedText: (beat: MulmoBeat, multiLingualData?: MulmoStudioMultiLingualData, lang?: string) => string;
+export declare const sleep: (milliseconds: number) => Promise<unknown>;

package/lib/utils/utils.js

@@ -52,3 +52,6 @@ export const localizedText = (beat, multiLingualData, lang) => {
     }
     return beat.text;
 };
+export const sleep = async (milliseconds) => {
+    return await new Promise((resolve) => setTimeout(resolve, milliseconds));
+};

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mulmocast",
-  "version": "0.0.9",
+  "version": "0.0.10",
   "description": "",
   "type": "module",
   "main": "lib/cli/bin.js",

package/scripts/templates/movie_prompts_template.json

@@ -0,0 +1,50 @@
+{
+  "$mulmocast": {
+    "version": "1.0",
+    "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]"
+    }
+  ],
+  "movieParams": {
+    "provider": "google"
+  },
+  "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.]",
+      "imagePrompt": "[IMAGE_PROMPT: A prompt for the image to be generated for this beat.]",
+      "moviePrompt": "[MOVIE_PROMPT: A movie prompt for that image.]"
+    },
+    {
+      "text": "[MAIN_CONCEPT: Define or explain the core concept/idea. This should be the central focus of your narrative. Keep it clear and accessible.]",
+      "imagePrompt": "[IMAGE_PROMPT: A prompt for the image to be generated for this beat.]",
+      "moviePrompt": "[MOVIE_PROMPT: A movie prompt for that image.]"
+    },
+    {
+      "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.]",
+      "imagePrompt": "[IMAGE_PROMPT: A prompt for the image to be generated for this beat.]",
+      "moviePrompt": "[MOVIE_PROMPT: A movie prompt for that image.]"
+    },
+    {
+      "text": "[SUPPORTING_DETAIL_2: Continue with more examples, deeper explanation, or different aspects of the topic if needed.]",
+      "imagePrompt": "[IMAGE_PROMPT: A prompt for the image to be generated for this beat.]",
+      "moviePrompt": "[MOVIE_PROMPT: A movie prompt for that image.]"
+    },
+    {
+      "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.]",
+      "imagePrompt": "[IMAGE_PROMPT: A prompt for the image to be generated for this beat.]",
+      "moviePrompt": "[MOVIE_PROMPT: A movie prompt for that image.]"
+    },
+    {
+      "text": "[CONCLUSION/IMPACT: Wrap up with the significance, implications, or key takeaway. Help the audience understand why this matters to them.]",
+      "imagePrompt": "[IMAGE_PROMPT: A prompt for the image to be generated for this beat.]",
+      "moviePrompt": "[MOVIE_PROMPT: A movie prompt for that image.]"
+    }
+  ]
+}