mulmocast 2.0.2 → 2.0.4

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/ffmpeg_utils.d.ts

@@ -18,3 +18,4 @@ export declare const ffmpegGetMediaDuration: (filePath: string) => Promise<{
 export declare const extractImageFromMovie: (movieFile: string, imagePath: string) => Promise<object>;
 export declare const trimMusic: (inputFile: string, startTime: number, duration: number) => Promise<Buffer>;
 export declare const createSilentAudio: (filePath: string, durationSec: number) => Promise<void>;
+export declare const pcmToMp3: (rawPcm: Buffer, sampleRate?: number) => Promise<Buffer>;

package/lib/utils/ffmpeg_utils.js

@@ -2,6 +2,7 @@ import ffmpeg from "fluent-ffmpeg";
 import { GraphAILogger } from "graphai";
 import { isFile } from "./file.js";
 import fs from "fs";
+import { Readable, PassThrough } from "node:stream";
 export const setFfmpegPath = (ffmpegPath) => {
     ffmpeg.setFfmpegPath(ffmpegPath);
 };
@@ -144,3 +145,25 @@ export const createSilentAudio = (filePath, durationSec) => {
             .run();
     });
 };
+export const pcmToMp3 = (rawPcm, sampleRate = 24000) => {
+    return new Promise((resolve, reject) => {
+        const inputStream = new Readable({
+            read() {
+                this.push(rawPcm);
+                this.push(null);
+            },
+        });
+        const outputChunks = [];
+        const outputStream = new PassThrough();
+        outputStream.on("data", (chunk) => outputChunks.push(chunk));
+        outputStream.on("end", () => resolve(Buffer.concat(outputChunks)));
+        outputStream.on("error", reject);
+        ffmpeg(inputStream)
+            .inputFormat("s16le")
+            .inputOptions([`-ar ${sampleRate}`, "-ac 1"])
+            .audioCodec("libmp3lame")
+            .format("mp3")
+            .on("error", reject)
+            .pipe(outputStream);
+    });
+};

package/lib/utils/provider2agent.d.ts

@@ -16,6 +16,12 @@ export declare const provider2TTSAgent: {
         hasLimitedConcurrency: boolean;
         keyName: string;
     };
+    gemini: {
+        agentName: string;
+        hasLimitedConcurrency: boolean;
+        defaultVoice: string;
+        keyName: string;
+    };
     elevenlabs: {
         agentName: string;
         hasLimitedConcurrency: boolean;
@@ -76,7 +82,10 @@ 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

@@ -17,6 +17,12 @@ export const provider2TTSAgent = {
         hasLimitedConcurrency: false,
         keyName: "GEMINI_API_KEY",
     },
+    gemini: {
+        agentName: "ttsGeminiAgent",
+        hasLimitedConcurrency: false,
+        defaultVoice: "Kore",
+        keyName: "GEMINI_API_KEY",
+    },
     elevenlabs: {
         agentName: "ttsElevenlabsAgent",
         hasLimitedConcurrency: true,
@@ -42,8 +48,8 @@ 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 +83,8 @@ 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 +150,12 @@ 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 +165,13 @@ 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/lib/utils/utils.js

@@ -73,6 +73,9 @@ export const settings2GraphAIConfig = (settings, env) => {
         ttsNijivoiceAgent: {
             apiKey: getKey("TTS", "NIJIVOICE_API_KEY"),
         },
+        ttsGeminiAgent: {
+            apiKey: getKey("TTS", "GEMINI_API_KEY"),
+        },
         ttsElevenlabsAgent: {
             apiKey: getKey("TTS", "ELEVENLABS_API_KEY"),
         },

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mulmocast",
-  "version": "2.0.2",
+  "version": "2.0.4",
   "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",
@@ -95,14 +95,14 @@
     "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"
+    "zod": "^4.1.13"
   },
   "devDependencies": {
     "@receptron/test_utils": "^2.0.3",
@@ -117,7 +117,7 @@
     "prettier": "^3.6.2",
     "tsx": "^4.20.6",
     "typescript": "^5.9.3",
-    "typescript-eslint": "^8.46.4"
+    "typescript-eslint": "^8.48.0"
   },
   "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_audio_gemini.json

@@ -0,0 +1,67 @@
+{
+  "$mulmocast": {
+    "version": "1.1"
+  },
+  "lang": "en",
+  "title": "Audio Instructions Test",
+  "speechParams": {
+    "speakers": {
+      "Presenter": {
+        "provider": "gemini",
+        "voiceId": "Kore"
+      },
+      "Presenter2": {
+        "provider": "gemini",
+        "voiceId": "Puck"
+      }
+    }
+  },
+  "beats": [
+    {
+      "speaker": "Presenter",
+      "text": "Hello, I'm a presenter. I have no instructions.",
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Presenter"
+        }
+      }
+    },
+    {
+      "speaker": "Presenter2",
+      "text": "Hello, I'm a presenter 2. My instructions are 'Speak in a cheerful and positive tone'.",
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Presenter 2"
+        }
+      }
+    },
+    {
+      "speaker": "Presenter",
+      "text": "Hello, I'm a presenter. I have a British English instruction.",
+      "speechOptions": {
+        "instruction": "Speak in British English."
+      },
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Presenter with British English instruction"
+        }
+      }
+    },
+    {
+      "speaker": "Presenter",
+      "text": "Hello, I'm a presenter. I have a whisper instruction.",
+      "speechOptions": {
+        "instruction": "Whisper softly, like a pillow talk."
+      },
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Presenter with whisper instruction"
+        }
+      }
+    }
+  ]
+}

package/scripts/test/test_audio_gemini.json~

@@ -0,0 +1,67 @@
+{
+  "$mulmocast": {
+    "version": "1.1"
+  },
+  "lang": "en",
+  "title": "Audio Instructions Test",
+  "speechParams": {
+    "speakers": {
+      "Presenter": {
+        "provider": "google",
+        "voiceId": "ja-JP-Standard-A"
+      },
+      "Presenter2": {
+        "provider": "google",
+        "voiceId": "ja-JP-Standard-B"
+      }
+    }
+  },
+  "beats": [
+    {
+      "speaker": "Presenter",
+      "text": "Hello, I'm a presenter. I have no instructions.",
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Presenter"
+        }
+      }
+    },
+    {
+      "speaker": "Presenter2",
+      "text": "Hello, I'm a presenter 2. My instructions are 'Speak in a cheerful and positive tone'.",
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Presenter 2"
+        }
+      }
+    },
+    {
+      "speaker": "Presenter",
+      "text": "Hello, I'm a presenter. I have a British English instruction.",
+      "speechOptions": {
+        "instruction": "Speak in British English."
+      },
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Presenter with British English instruction"
+        }
+      }
+    },
+    {
+      "speaker": "Presenter",
+      "text": "Hello, I'm a presenter. I have a whisper instruction.",
+      "speechOptions": {
+        "instruction": "Whisper softly, like a pillow talk."
+      },
+      "image": {
+        "type": "textSlide",
+        "slide": {
+          "title": "Presenter with whisper instruction"
+        }
+      }
+    }
+  ]
+}