mulmocast 1.1.5 → 1.1.6

package/lib/actions/audio.js

@@ -10,7 +10,7 @@ import ttsElevenlabsAgent from "../agents/tts_elevenlabs_agent.js";
 import { fileWriteAgent } from "@graphai/vanilla_node_agents";
 import { MulmoPresentationStyleMethods } from "../methods/index.js";
 import { text2SpeechProviderSchema } from "../types/index.js";
-import { fileCacheAgentFilter } from "../utils/filters.js";
+import { fileCacheAgentFilter, nijovoiceTextAgentFilter } from "../utils/filters.js";
 import { getAudioArtifactFilePath, getAudioFilePath, getOutputStudioFilePath, resolveDirPath, defaultBGMPath, mkdir, writingMessage } from "../utils/file.js";
 import { text2hash, localizedText, settings2GraphAIConfig } from "../utils/utils.js";
 import { provider2TTSAgent } from "../utils/provider2agent.js";
@@ -58,6 +58,8 @@ const preprocessor = (namedInputs) => {
         voiceId,
         speechOptions,
         model,
+        provider,
+        lang,
         audioPath,
         studioBeat,
         needsTTS,
@@ -84,6 +86,8 @@ const graph_tts = {
             agent: ":preprocessor.ttsAgent",
             inputs: {
                 text: ":preprocessor.text",
+                provider: ":preprocessor.provider",
+                lang: ":preprocessor.lang",
                 cache: {
                     force: [":context.force"],
                     file: ":preprocessor.audioPath",
@@ -173,6 +177,11 @@ const agentFilters = [
         agent: fileCacheAgentFilter,
         nodeIds: ["tts"],
     },
+    {
+        name: "nijovoiceTextAgentFilter",
+        agent: nijovoiceTextAgentFilter,
+        nodeIds: ["tts"],
+    },
 ];
 const getConcurrency = (context) => {
     // Check if any speaker uses nijivoice or elevenlabs (providers that require concurrency = 1)

package/lib/actions/image_agents.d.ts

@@ -23,10 +23,7 @@ export declare const imagePreprocessAgent: (namedInputs: {
     };
     lipSyncFile?: string;
     lipSyncModel?: string;
-    lipSyncAgentInfo?: {
-        agentName: string;
-        defaultModel: string;
-    };
+    lipSyncAgentName?: string;
     audioFile?: string;
     beatDuration?: number;
     htmlPrompt?: undefined;
@@ -61,10 +58,7 @@ export declare const imagePreprocessAgent: (namedInputs: {
     };
     lipSyncFile?: string;
     lipSyncModel?: string;
-    lipSyncAgentInfo?: {
-        agentName: string;
-        defaultModel: string;
-    };
+    lipSyncAgentName?: string;
     audioFile?: string;
     beatDuration?: number;
     htmlPrompt?: undefined;
@@ -102,10 +96,7 @@ export declare const imagePreprocessAgent: (namedInputs: {
     };
     lipSyncFile?: string;
     lipSyncModel?: string;
-    lipSyncAgentInfo?: {
-        agentName: string;
-        defaultModel: string;
-    };
+    lipSyncAgentName?: string;
     audioFile?: string;
     beatDuration?: number;
     htmlPrompt?: undefined;

package/lib/actions/image_agents.js

@@ -25,16 +25,20 @@ export const imagePreprocessAgent = async (namedInputs) => {
         movieFile: beat.moviePrompt ? moviePaths.movieFile : undefined,
         beatDuration: beat.duration ?? studioBeat?.duration,
     };
-    if (beat.soundEffectPrompt) {
-        returnValue.soundEffectAgentInfo = MulmoPresentationStyleMethods.getSoundEffectAgentInfo(context.presentationStyle, beat);
-        returnValue.soundEffectModel =
-            beat.soundEffectParams?.model ?? context.presentationStyle.soundEffectParams?.model ?? returnValue.soundEffectAgentInfo.defaultModel;
-        returnValue.soundEffectFile = moviePaths.soundEffectFile;
-        returnValue.soundEffectPrompt = beat.soundEffectPrompt;
+    const isMovie = Boolean(beat.moviePrompt || beat?.image?.type === "movie");
+    if (isMovie) {
+        if (beat.soundEffectPrompt) {
+            returnValue.soundEffectAgentInfo = MulmoPresentationStyleMethods.getSoundEffectAgentInfo(context.presentationStyle, beat);
+            returnValue.soundEffectModel =
+                beat.soundEffectParams?.model ?? context.presentationStyle.soundEffectParams?.model ?? returnValue.soundEffectAgentInfo.defaultModel;
+            returnValue.soundEffectFile = moviePaths.soundEffectFile;
+            returnValue.soundEffectPrompt = beat.soundEffectPrompt;
+        }
     }
     if (beat.enableLipSync) {
-        returnValue.lipSyncAgentInfo = MulmoPresentationStyleMethods.getLipSyncAgentInfo(context.presentationStyle, beat);
-        returnValue.lipSyncModel = beat.lipSyncParams?.model ?? context.presentationStyle.lipSyncParams?.model ?? returnValue.lipSyncAgentInfo.defaultModel;
+        const lipSyncAgentInfo = MulmoPresentationStyleMethods.getLipSyncAgentInfo(context.presentationStyle, beat);
+        returnValue.lipSyncAgentName = lipSyncAgentInfo.agentName;
+        returnValue.lipSyncModel = beat.lipSyncParams?.model ?? context.presentationStyle.lipSyncParams?.model ?? lipSyncAgentInfo.defaultModel;
         returnValue.lipSyncFile = moviePaths.lipSyncFile;
         // Audio file will be set from the beat's audio file when available
         returnValue.audioFile = studioBeat?.audioFile;

package/lib/actions/images.js

@@ -218,10 +218,11 @@ const beat_graph_data = {
         },
         lipSyncGenerator: {
             if: ":beat.enableLipSync",
-            agent: ":preprocessor.lipSyncAgentInfo.agentName",
+            agent: ":preprocessor.lipSyncAgentName",
             inputs: {
                 onComplete: [":soundEffectGenerator"], // to wait for soundEffectGenerator to finish
                 movieFile: ":preprocessor.movieFile",
+                imageFile: ":preprocessor.referenceImageForMovie",
                 audioFile: ":preprocessor.audioFile",
                 lipSyncFile: ":preprocessor.lipSyncFile",
                 params: {

package/lib/actions/translate.js

@@ -1,9 +1,9 @@
 import "dotenv/config";
-import { GraphAI, assert } from "graphai";
+import { GraphAI, assert, isNull } from "graphai";
 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 { recursiveSplitJa } from "../utils/string.js";
 import { settings2GraphAIConfig } from "../utils/utils.js";
 import { getOutputMultilingualFilePath, mkdir, writingMessage } from "../utils/file.js";
 import { translateSystemPrompt, translatePrompts } from "../utils/prompt.js";
@@ -13,17 +13,9 @@ const translateGraph = {
     version: 0.5,
     nodes: {
         context: {},
-        defaultLang: {},
         outDirPath: {},
         outputMultilingualFilePath: {},
-        lang: {
-            agent: "stringUpdateTextAgent",
-            inputs: {
-                newText: ":context.studio.script.lang",
-                oldText: ":defaultLang",
-            },
-        },
-        targetLangs: {}, // TODO
+        targetLangs: {},
         mergeStudioResult: {
             isResult: true,
             agent: "mergeObjectAgent",
@@ -37,7 +29,6 @@ const translateGraph = {
                 targetLangs: ":targetLangs",
                 context: ":context",
                 rows: ":context.studio.script.beats",
-                lang: ":lang",
             },
             params: {
                 rowKey: "beat",
@@ -62,7 +53,7 @@ const translateGraph = {
                             beat: ":beat",
                             multiLingual: ":multiLingual",
                             rows: ":targetLangs",
-                            lang: ":lang.text",
+                            lang: ":context.studio.script.lang",
                             context: ":context",
                             beatIndex: ":__mapIndex",
                         },
@@ -120,17 +111,11 @@ const translateGraph = {
                                 },
                                 ttsTexts: {
                                     agent: (namedInputs) => {
-                                        const { localizedText, targetLang } = namedInputs;
+                                        const { localizedText } = namedInputs;
                                         // cache
                                         if (localizedText.ttsTexts) {
                                             return localizedText;
                                         }
-                                        if (targetLang === "ja") {
-                                            return {
-                                                ...localizedText,
-                                                ttsTexts: localizedText?.texts?.map((text) => replacePairsJa(text, replacementsJa)),
-                                            };
-                                        }
                                         return {
                                             ...localizedText,
                                             ttsTexts: localizedText.texts,
@@ -180,18 +165,14 @@ const localizedTextCacheAgentFilter = async (context, next) => {
     if (!beat.text) {
         return { text: "" };
     }
-    // The original text is unchanged and the target language text is present
-    if (multiLingual.multiLingualTexts &&
-        multiLingual.multiLingualTexts[lang] &&
-        multiLingual.multiLingualTexts[lang].text === beat.text &&
-        multiLingual.multiLingualTexts[targetLang] &&
-        multiLingual.multiLingualTexts[targetLang].text) {
-        return { text: multiLingual.multiLingualTexts[targetLang].text };
-    }
     // same language
     if (targetLang === lang) {
         return { text: beat.text };
     }
+    // The original text is unchanged and the target language text is present
+    if (multiLingual.multiLingualTexts?.[lang]?.text === beat.text && multiLingual.multiLingualTexts[targetLang]?.text) {
+        return { text: multiLingual.multiLingualTexts[targetLang].text };
+    }
     try {
         MulmoStudioContextMethods.setBeatSessionState(mulmoContext, "multiLingual", beatIndex, true);
         return await next(context);
@@ -207,8 +188,6 @@ const agentFilters = [
         nodeIds: ["localizedTexts"],
     },
 ];
-const defaultLang = "en";
-const targetLangs = ["ja", "en"];
 export const translate = async (context, args) => {
     const { settings, callbacks } = args ?? {};
     try {
@@ -217,11 +196,14 @@ export const translate = async (context, args) => {
         const outDirPath = MulmoStudioContextMethods.getOutDirPath(context);
         const outputMultilingualFilePath = getOutputMultilingualFilePath(outDirPath, fileName);
         mkdir(outDirPath);
+        const langs = (context.multiLingual ?? []).map((x) => Object.keys(x.multiLingualTexts)).flat(); // existing langs in multiLingual
+        const targetLangs = [
+            ...new Set([context.studio.script.lang, langs, context.lang, context.studio.script.captionParams?.lang].flat().filter((x) => !isNull(x))),
+        ];
         const config = settings2GraphAIConfig(settings, process.env);
         assert(!!config?.openAIAgent?.apiKey, "The OPENAI_API_KEY environment variable is missing or empty");
         const graph = new GraphAI(translateGraph, { ...vanillaAgents, fileWriteAgent, openAIAgent }, { agentFilters, config });
         graph.injectValue("context", context);
-        graph.injectValue("defaultLang", defaultLang);
         graph.injectValue("targetLangs", targetLangs);
         graph.injectValue("outDirPath", outDirPath);
         graph.injectValue("outputMultilingualFilePath", outputMultilingualFilePath);

package/lib/agents/lipsync_replicate_agent.js

@@ -3,7 +3,7 @@ import { GraphAILogger } from "graphai";
 import Replicate from "replicate";
 import { provider2LipSyncAgent } from "../utils/provider2agent.js";
 export const lipSyncReplicateAgent = async ({ namedInputs, params, config, }) => {
-    const { movieFile, audioFile } = namedInputs;
+    const { movieFile, audioFile, imageFile } = namedInputs;
     const apiKey = config?.apiKey;
     const model = params.model ?? provider2LipSyncAgent.replicate.defaultModel;
     if (!apiKey) {
@@ -12,10 +12,12 @@ export const lipSyncReplicateAgent = async ({ namedInputs, params, config, }) =>
     const replicate = new Replicate({
         auth: apiKey,
     });
-    const videoBuffer = readFileSync(movieFile);
+    const videoBuffer = movieFile ? readFileSync(movieFile) : undefined;
     const audioBuffer = readFileSync(audioFile);
-    const videoUri = `data:video/quicktime;base64,${videoBuffer.toString("base64")}`;
+    const imageBuffer = imageFile ? readFileSync(imageFile) : undefined;
+    const videoUri = videoBuffer ? `data:video/quicktime;base64,${videoBuffer.toString("base64")}` : undefined;
     const audioUri = `data:audio/wav;base64,${audioBuffer.toString("base64")}`;
+    const imageUri = imageBuffer ? `data:image/png;base64,${imageBuffer.toString("base64")}` : undefined;
     const input = {
         video: undefined,
         video_input: undefined,
@@ -23,6 +25,7 @@ export const lipSyncReplicateAgent = async ({ namedInputs, params, config, }) =>
         audio: undefined,
         audio_input: undefined,
         audio_file: undefined,
+        image: undefined,
     };
     const modelParams = provider2LipSyncAgent.replicate.modelParams[model];
     if (!modelParams) {
@@ -30,12 +33,16 @@ export const lipSyncReplicateAgent = async ({ namedInputs, params, config, }) =>
     }
     const videoParam = modelParams.video;
     const audioParam = modelParams.audio;
+    const imageParam = modelParams.image;
     if (videoParam === "video" || videoParam === "video_input" || videoParam === "video_url") {
         input[videoParam] = videoUri;
     }
     if (audioParam === "audio" || audioParam === "audio_input" || audioParam === "audio_file") {
         input[audioParam] = audioUri;
     }
+    if (imageParam === "image") {
+        input[imageParam] = imageUri;
+    }
     const model_identifier = provider2LipSyncAgent.replicate.modelParams[model]?.identifier ?? model;
     try {
         const output = await replicate.run(model_identifier, {

package/lib/cli/commands/audio/handler.js

@@ -5,6 +5,6 @@ export const handler = async (argv) => {
     if (!context) {
         process.exit(1);
     }
-    await runTranslateIfNeeded(context, argv);
+    await runTranslateIfNeeded(context);
     await audio(context);
 };

package/lib/cli/commands/image/handler.js

@@ -5,6 +5,6 @@ export const handler = async (argv) => {
     if (!context) {
         process.exit(1);
     }
-    await runTranslateIfNeeded(context, argv);
+    await runTranslateIfNeeded(context);
     await images(context);
 };

package/lib/cli/commands/movie/handler.js

@@ -5,6 +5,6 @@ export const handler = async (argv) => {
     if (!context) {
         process.exit(1);
     }
-    await runTranslateIfNeeded(context, argv);
+    await runTranslateIfNeeded(context, true);
     await audio(context).then(images).then(captions).then(movie);
 };

package/lib/cli/commands/pdf/handler.js

@@ -5,7 +5,7 @@ export const handler = async (argv) => {
     if (!context) {
         process.exit(1);
     }
-    await runTranslateIfNeeded(context, argv);
+    await runTranslateIfNeeded(context);
     await images(context);
     await pdf(context, argv.pdf_mode, argv.pdf_size);
 };

package/lib/cli/helpers.d.ts

@@ -1,9 +1,6 @@
 import type { CliArgs } from "../types/cli_types.js";
 import { FileObject, InitOptions, MulmoStudioContext } from "../types/index.js";
-export declare const runTranslateIfNeeded: (context: MulmoStudioContext, argv: {
-    l?: string;
-    c?: string;
-}) => Promise<void>;
+export declare const runTranslateIfNeeded: (context: MulmoStudioContext, includeCaption?: boolean) => Promise<void>;
 export declare const setGraphAILogger: (verbose: boolean | undefined, logValues?: Record<string, unknown>) => void;
 export declare const getFileObject: (args: {
     basedir?: string;

package/lib/cli/helpers.js

@@ -5,10 +5,11 @@ import clipboardy from "clipboardy";
 import { getBaseDirPath, getFullPath, getOutputStudioFilePath, resolveDirPath, mkdir, getOutputMultilingualFilePath, generateTimestampedFileName, } from "../utils/file.js";
 import { isHttp } from "../utils/utils.js";
 import { outDirName, imageDirName, audioDirName } from "../utils/const.js";
+import { MulmoStudioContextMethods } from "../methods/mulmo_studio_context.js";
 import { translate } from "../actions/translate.js";
 import { initializeContextFromFiles } from "../utils/context.js";
-export const runTranslateIfNeeded = async (context, argv) => {
-    if (argv.l || context.studio.script.captionParams?.lang) {
+export const runTranslateIfNeeded = async (context, includeCaption = false) => {
+    if (MulmoStudioContextMethods.needTranslate(context, includeCaption)) {
         GraphAILogger.log("run translate");
         await translate(context);
     }

package/lib/mcp/server.js

@@ -104,7 +104,7 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
             throw new Error("Failed to initialize context from MulmoScript");
         }
         // Run translation if needed
-        await runTranslateIfNeeded(context, argv);
+        await runTranslateIfNeeded(context);
         // Execute the requested command
         switch (cmd) {
             case "movie":

package/lib/methods/mulmo_presentation_style.d.ts

@@ -36,9 +36,10 @@ export declare const MulmoPresentationStyleMethods: {
         defaultModel: import("../utils/provider2agent.js").ReplicateModel;
         models: import("../utils/provider2agent.js").ReplicateModel[];
         modelParams: Record<import("../utils/provider2agent.js").ReplicateModel, {
-            identifier?: `${string}/${string}:${string}`;
-            video: string;
+            identifier?: `${string}/${string}:${string}` | `${string}/${string}`;
+            video?: string;
             audio: string;
+            image?: string;
         }>;
     };
     getConcurrency(presentationStyle: MulmoPresentationStyle): 4 | 16;

package/lib/methods/mulmo_script.js

@@ -18,6 +18,9 @@ const validators = [{ from: "1.0", to: "1.1", validator: validate_1_0 }];
 export const MulmoScriptMethods = {
     // eslint-disable-next-line @typescript-eslint/no-explicit-any
     validate(script) {
+        const version = script.$mulmocast.version;
+        // lang was optional in 1.0 and 1.1
+        const defaultLang = version === "1.0" || version === "1.1" ? { lang: "en" } : {};
         const validatedScript = validators.reduce((acc, validator) => {
             if (acc.$mulmocast.version === validator.from) {
                 const validated = validator.validator(acc);
@@ -25,7 +28,7 @@ export const MulmoScriptMethods = {
                 return validated;
             }
             return acc;
-        }, script);
+        }, { ...defaultLang, ...script });
         return mulmoScriptSchema.parse(validatedScript);
     },
 };

package/lib/methods/mulmo_studio_context.d.ts

@@ -11,4 +11,5 @@ export declare const MulmoStudioContextMethods: {
     getCaption(context: MulmoStudioContext): string | undefined;
     setSessionState(context: MulmoStudioContext, sessionType: SessionType, value: boolean): void;
     setBeatSessionState(context: MulmoStudioContext, sessionType: BeatSessionType, index: number, value: boolean): void;
+    needTranslate(context: MulmoStudioContext, includeCaption?: boolean): boolean | "" | undefined;
 };

package/lib/methods/mulmo_studio_context.js

@@ -63,4 +63,12 @@ export const MulmoStudioContextMethods = {
         }
         notifyBeatStateChange(context, sessionType, index);
     },
+    needTranslate(context, includeCaption = false) {
+        // context.studio.script.lang = defaultLang, context.lang = targetLanguage.
+        if (includeCaption) {
+            return (context.studio.script.lang !== context.lang ||
+                (context.studio.script.captionParams?.lang && context.studio.script.lang !== context.studio.script.captionParams?.lang));
+        }
+        return context.studio.script.lang !== context.lang;
+    },
 };

package/lib/types/agent.d.ts

@@ -77,6 +77,7 @@ export type LipSyncAgentInputs = {
     lipSyncFile: string;
     movieFile: string;
     audioFile: string;
+    imageFile: string;
 };
 export type GoogleMovieAgentConfig = GoogleImageAgentConfig;
 export type ReplicateMovieAgentConfig = AgentConfig;

package/lib/types/schema.d.ts

@@ -3861,7 +3861,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         title?: string | undefined;
         description?: string | undefined;
     }>, "many">>;
-    lang: z.ZodOptional<z.ZodString>;
+    lang: z.ZodString;
     beats: z.ZodArray<z.ZodObject<{
         speaker: z.ZodOptional<z.ZodString>;
         text: z.ZodDefault<z.ZodString>;
@@ -5036,6 +5036,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
     imagePath: z.ZodOptional<z.ZodString>;
     __test_invalid__: z.ZodOptional<z.ZodBoolean>;
 }, "strict", z.ZodTypeAny, {
+    lang: string;
     imageParams: {
         provider: string;
         model?: string | undefined;
@@ -5332,7 +5333,6 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         } | undefined;
         enableLipSync?: boolean | undefined;
     }[];
-    lang?: string | undefined;
     title?: string | undefined;
     description?: string | undefined;
     lipSyncParams?: {
@@ -5359,6 +5359,7 @@ export declare const mulmoScriptSchema: z.ZodObject<{
     imagePath?: string | undefined;
     __test_invalid__?: boolean | undefined;
 }, {
+    lang: string;
     $mulmocast: {
         version: "1.1";
         credit?: "closing" | undefined;
@@ -5564,7 +5565,6 @@ export declare const mulmoScriptSchema: z.ZodObject<{
         } | undefined;
         enableLipSync?: boolean | undefined;
     }[];
-    lang?: string | undefined;
     title?: string | undefined;
     description?: string | undefined;
     imageParams?: {
@@ -6393,7 +6393,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             title?: string | undefined;
             description?: string | undefined;
         }>, "many">>;
-        lang: z.ZodOptional<z.ZodString>;
+        lang: z.ZodString;
         beats: z.ZodArray<z.ZodObject<{
             speaker: z.ZodOptional<z.ZodString>;
             text: z.ZodDefault<z.ZodString>;
@@ -7568,6 +7568,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         imagePath: z.ZodOptional<z.ZodString>;
         __test_invalid__: z.ZodOptional<z.ZodBoolean>;
     }, "strict", z.ZodTypeAny, {
+        lang: string;
         imageParams: {
             provider: string;
             model?: string | undefined;
@@ -7864,7 +7865,6 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             enableLipSync?: boolean | undefined;
         }[];
-        lang?: string | undefined;
         title?: string | undefined;
         description?: string | undefined;
         lipSyncParams?: {
@@ -7891,6 +7891,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         imagePath?: string | undefined;
         __test_invalid__?: boolean | undefined;
     }, {
+        lang: string;
         $mulmocast: {
             version: "1.1";
             credit?: "closing" | undefined;
@@ -8096,7 +8097,6 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             enableLipSync?: boolean | undefined;
         }[];
-        lang?: string | undefined;
         title?: string | undefined;
         description?: string | undefined;
         imageParams?: {
@@ -8275,6 +8275,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         captionFile?: string | undefined;
     }[];
     script: {
+        lang: string;
         imageParams: {
             provider: string;
             model?: string | undefined;
@@ -8571,7 +8572,6 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             enableLipSync?: boolean | undefined;
         }[];
-        lang?: string | undefined;
         title?: string | undefined;
         description?: string | undefined;
         lipSyncParams?: {
@@ -8616,6 +8616,7 @@ export declare const mulmoStudioSchema: z.ZodObject<{
         captionFile?: string | undefined;
     }[];
     script: {
+        lang: string;
         $mulmocast: {
             version: "1.1";
             credit?: "closing" | undefined;
@@ -8821,7 +8822,6 @@ export declare const mulmoStudioSchema: z.ZodObject<{
             } | undefined;
             enableLipSync?: boolean | undefined;
         }[];
-        lang?: string | undefined;
         title?: string | undefined;
         description?: string | undefined;
         imageParams?: {

package/lib/types/schema.js

@@ -371,7 +371,7 @@ export const mulmoScriptSchema = mulmoPresentationStyleSchema
     title: z.string().optional(),
     description: z.string().optional(),
     references: z.array(mulmoReferenceSchema).optional(),
-    lang: langSchema.optional(), // default "en"
+    lang: langSchema, // required (default WAS "en")
     beats: z.array(mulmoBeatSchema).min(1),
     // TODO: Delete it later
     imagePath: z.string().optional(), // for keynote images movie ??

package/lib/types/type.d.ts

@@ -53,7 +53,7 @@ export type FileDirs = {
 export type MulmoStudioContext = {
     fileDirs: FileDirs;
     studio: MulmoStudio;
-    lang?: string;
+    lang: string;
     force: boolean;
     sessionState: MulmoSessionState;
     presentationStyle: MulmoPresentationStyle;

package/lib/utils/const.js

@@ -4,7 +4,7 @@ export const imageDirName = "images";
 export const cacheDirName = "cache";
 export const pdf_modes = ["slide", "talk", "handout"];
 export const pdf_sizes = ["letter", "a4"];
-export const languages = ["en", "ja"];
+export const languages = ["en", "ja", "fr", "es"];
 export const storyToScriptGenerateMode = {
     stepWise: "step_wise",
     oneStep: "one_step",