@doufunao123/asset-gateway 0.21.0 → 0.22.0

package/README.md

@@ -59,6 +59,9 @@ asset-gateway generate image --prompt "a cat" --size 1024x1024
 asset-gateway generate image --prompt "icon" --transparent --provider flux
 asset-gateway generate video --prompt "ocean waves"
 asset-gateway generate sfx --prompt "epic battle impact" --duration 3
+asset-gateway generate tts --prompt "（开心）今天天气真好！" --voice 冰糖
+asset-gateway voice design --voice-prompt "warm narrator" --preview-text "Welcome." --name narrator --output ./narrator.wav
+asset-gateway voice clone --audio ./sample.wav --preview-text "Welcome back." --name clone --output ./clone.wav
 asset-gateway generate character --prompt "medieval knight" --format fbx --pbr
 asset-gateway generate prop --prompt "ornate treasure chest" --polycount 5000
 asset-gateway generate model --image https://example.com/ref.png --ai-model latest
@@ -86,6 +89,20 @@ asset-gateway describe
 asset-gateway describe generate
 ```
 
+Programmatic TTS uses the SDK directly. MiMo voices include `冰糖`, `茉莉`, `苏打`, `白桦`, `Mia`, `Chloe`, `Milo`, and `Dean`; prompts may include audio tags such as `（开心）`, `(唱歌)`, or `[whisper]`.
+
+```ts
+await forge.tts("（开心）今天天气真好！", { voice: "冰糖" });
+```
+
+Voice design and clone are also available from the CLI:
+
+```bash
+asset-gateway voice design --voice-prompt "warm documentary narrator" --preview-text "Welcome to AssetForge." --name narrator --output ./narrator.wav
+asset-gateway voice clone --audio ./sample.wav --preview-text "Welcome back." --name clone --output ./clone.wav
+asset-gateway voice list --type vc
+```
+
 ## Output
 
 JSON by default. Use `--human` for readable output, `--fields` to filter:

package/dist/index.js

@@ -1,7 +1,7 @@
 #!/usr/bin/env node
 
 // src/index.ts
-import { Command as Command10 } from "commander";
+import { Command as Command11 } from "commander";
 
 // src/commands/auth.ts
 import { existsSync as existsSync2, unlinkSync } from "fs";
@@ -303,11 +303,11 @@ var SCHEMAS = {
         }
       },
       tts: {
-        description: "Text-to-speech via Gemini 3.1 Flash TTS",
+        description: "Text-to-speech via MiMo v2.5 TTS",
         params: {
           "--prompt": { type: "string", required: true },
-          "--voice": { type: "string", description: "Prebuilt voice name (default: Kore)" },
-          "--speakers": { type: "string", description: `Multi-speaker config JSON, e.g. '{"Name1":"Puck","Name2":"Kore"}'` },
+          "--voice": { type: "string", description: "MiMo prebuilt voice name (default: server config, usually Mia)" },
+          "--context": { type: "string", description: "Natural-language style or director instruction" },
           "--output-dir": { type: "string", default: "." }
         }
       },
@@ -513,6 +513,36 @@ var SCHEMAS = {
       health: { description: "Health check", params: { "[name]": { type: "string", required: false } } }
     }
   },
+  voice: {
+    description: "Design, clone, list, and delete MiMo voices",
+    subcommands: {
+      design: {
+        description: "Generate a MiMo voice from a text description",
+        params: {
+          "--voice-prompt": { type: "string", required: true },
+          "--preview-text": { type: "string", required: true },
+          "--style": { type: "string" },
+          "--name": { type: "string" },
+          "--save-as": { type: "string" },
+          "--output": { type: "string", description: "Write preview WAV to this path" }
+        }
+      },
+      clone: {
+        description: "Clone a MiMo voice from an mp3/wav sample",
+        params: {
+          "--audio": { type: "string", required: true, description: "Voice sample mp3/wav file or data URL" },
+          "--preview-text": { type: "string", required: true },
+          "--audio-mime": { type: "string" },
+          "--style": { type: "string" },
+          "--name": { type: "string" },
+          "--save-as": { type: "string" },
+          "--output": { type: "string", description: "Write preview WAV to this path" }
+        }
+      },
+      list: { description: "List saved voices", params: { "--type": { type: "string", description: "vc | vd" } } },
+      delete: { description: "Delete a saved voice", params: { "<voice-id>": { type: "string", required: true }, "--type": { type: "string", description: "vc | vd" } } }
+    }
+  },
   job: {
     description: "Async job history",
     subcommands: {
@@ -576,7 +606,7 @@ function inferExtension(assetType) {
     audio: "mp3",
     sfx: "mp3",
     music: "mp3",
-    tts: "mp3",
+    tts: "wav",
     video: "mp4",
     model3d: "glb",
     character: "glb",
@@ -599,6 +629,8 @@ function inferExtFromResult(result) {
     "image/jpeg": "jpg",
     "video/mp4": "mp4",
     "audio/mpeg": "mp3",
+    "audio/wav": "wav",
+    "audio/x-wav": "wav",
     "model/gltf-binary": "glb",
     "model/gltf+json": "gltf",
     "model/stl": "stl",
@@ -743,9 +775,12 @@ function createGenerateCommand() {
     new Command3("video").description("Generate a video from a text prompt (or image-to-video with --input)").requiredOption("--prompt <text>", "Video description prompt").option("--provider <id>", "Provider to use").option("--input <url>", "Reference image URL for image-to-video (Grok)").option("--output-dir <dir>", "Directory to save output", ".").action(async function(options) {
       try {
         const ctx = createContext(this);
-        const data = await ctx.client.video(options.prompt, {
+        const requestOptions = {
           provider: options.provider,
           input: options.input
+        };
+        const data = await ctx.client.video(options.prompt, {
+          ...requestOptions
         });
         const localPath = await saveOutput(data, "video", options.outputDir);
         if (localPath) data.local_path = localPath;
@@ -841,11 +876,11 @@ function createGenerateCommand() {
     })
   );
   command.addCommand(
-    new Command3("tts").description("Text-to-speech via Gemini 3.1 Flash TTS").requiredOption("--prompt <text>", "Text to synthesize").option("--voice <name>", "Prebuilt voice name (default: Kore)").option("--speakers <json>", `Multi-speaker config JSON, e.g. '{"Name1":"Puck","Name2":"Kore"}'`).option("--output-dir <dir>", "Directory to save output", ".").action(async function(options) {
+    new Command3("tts").description("Text-to-speech via MiMo v2.5 TTS").requiredOption("--prompt <text>", "Text to synthesize").option("--voice <name>", "MiMo prebuilt voice name (default: server config, usually Mia)").option("--context <text>", "Natural-language style or director instruction").option("--output-dir <dir>", "Directory to save output", ".").action(async function(options) {
       try {
         const ctx = createContext(this);
         const params = {};
-        if (options.speakers) params.speakers = JSON.parse(options.speakers);
+        if (options.context) params.context = options.context;
         const data = await ctx.client.tts(options.prompt, {
           voice: options.voice,
           params: Object.keys(params).length > 0 ? toJsonObject(params) : void 0
@@ -1505,8 +1540,114 @@ async function readLocalFile(filePath) {
   }
 }
 
+// src/commands/voice.ts
+import { existsSync as existsSync6, mkdirSync as mkdirSync5, readFileSync as readFileSync5, writeFileSync as writeFileSync5 } from "fs";
+import { dirname as dirname3, extname as extname2 } from "path";
+import { Command as Command10 } from "commander";
+function createVoiceCommand() {
+  const command = new Command10("voice").description("Design, clone, list, and delete voices");
+  command.addCommand(
+    new Command10("design").description("Generate a MiMo voice from a text description").requiredOption("--voice-prompt <text>", "Voice description / style prompt").requiredOption("--preview-text <text>", "Text to synthesize for preview").option("--style <text>", "Additional director/style instruction").option("--name <name>", "Save generated voice with this name").option("--save-as <name>", "Alias for --name").option("--output <path>", "Write preview WAV to this path").action(async function(options) {
+      const ctx = createContext(this);
+      try {
+        const request = {
+          voice_prompt: options.voicePrompt,
+          preview_text: options.previewText,
+          style: options.style,
+          name: options.name,
+          save_as: options.saveAs
+        };
+        const data = await ctx.client.voice.design(request);
+        const localPath = writeWavOutput(data, options.output);
+        if (localPath) data.local_path = localPath;
+        printSuccess("voice.design", data, ctx);
+      } catch (error2) {
+        printError("voice.design", error2, ctx.human);
+      }
+    })
+  );
+  command.addCommand(
+    new Command10("clone").description("Clone a MiMo voice from an mp3/wav sample").requiredOption("--audio <path-or-data-url>", "Voice sample mp3/wav file or data URL").requiredOption("--preview-text <text>", "Text to synthesize for preview").option("--audio-mime <mime>", "Audio MIME type when --audio is raw base64").option("--style <text>", "Director/style instruction").option("--name <name>", "Save cloned voice with this name").option("--save-as <name>", "Alias for --name").option("--output <path>", "Write preview WAV to this path").action(async function(options) {
+      const ctx = createContext(this);
+      try {
+        const sample = readVoiceSample(options.audio, options.audioMime);
+        const request = {
+          ...sample,
+          preview_text: options.previewText,
+          style: options.style,
+          name: options.name,
+          save_as: options.saveAs
+        };
+        const data = await ctx.client.voice.clone(request);
+        const localPath = writeWavOutput(data, options.output);
+        if (localPath) data.local_path = localPath;
+        printSuccess("voice.clone", data, ctx);
+      } catch (error2) {
+        printError("voice.clone", error2, ctx.human);
+      }
+    })
+  );
+  command.addCommand(
+    new Command10("list").description("List saved voices").option("--type <type>", "Voice type: vc or vd").action(async function(options) {
+      const ctx = createContext(this);
+      try {
+        const data = await ctx.client.voice.list({ type: options.type });
+        printSuccess("voice.list", data, ctx);
+      } catch (error2) {
+        printError("voice.list", error2, ctx.human);
+      }
+    })
+  );
+  command.addCommand(
+    new Command10("delete").description("Delete a saved voice").argument("<voice-id>", "Voice ID").option("--type <type>", "Voice type: vc or vd").action(async function(voiceId, options) {
+      const ctx = createContext(this);
+      try {
+        const data = await ctx.client.voice.delete(voiceId, { type: options.type });
+        printSuccess("voice.delete", data, ctx);
+      } catch (error2) {
+        printError("voice.delete", error2, ctx.human);
+      }
+    })
+  );
+  return command;
+}
+function readVoiceSample(input, audioMime) {
+  if (input.startsWith("data:")) {
+    return { sample_data_url: input };
+  }
+  if (existsSync6(input)) {
+    return {
+      audio_base64: readFileSync5(input).toString("base64"),
+      audio_mime: audioMime ?? inferAudioMime(input)
+    };
+  }
+  return { audio_base64: input, audio_mime: audioMime ?? "audio/wav" };
+}
+function inferAudioMime(filePath) {
+  const ext = extname2(filePath).toLowerCase();
+  if (ext === ".mp3") return "audio/mpeg";
+  if (ext === ".wav") return "audio/wav";
+  return "audio/wav";
+}
+function writeWavOutput(data, outputPath) {
+  if (!outputPath) {
+    return null;
+  }
+  const raw = data.wav_base64 ?? data.output_data;
+  if (typeof raw !== "string" || !raw) {
+    return null;
+  }
+  mkdirSync5(dirname3(outputPath), { recursive: true });
+  writeFileSync5(outputPath, Buffer.from(stripDataUri2(raw), "base64"));
+  return outputPath;
+}
+function stripDataUri2(data) {
+  const idx = data.indexOf(";base64,");
+  return idx >= 0 ? data.slice(idx + 8) : data;
+}
+
 // src/index.ts
-var program = new Command10().name("asset-gateway").description("Universal asset generation gateway CLI").version(CLI_VERSION).option(
+var program = new Command11().name("asset-gateway").description("Universal asset generation gateway CLI").version(CLI_VERSION).option(
   "--gateway-url <url>",
   `Gateway URL (default: $ASSET_GATEWAY_URL, auth config, or ${DEFAULT_GATEWAY_URL})`
 ).option("--token <token>", "API token for authentication").option("--human", "Human-readable output instead of JSON").option("--fields <fields>", "Comma-separated list of output fields");
@@ -1518,5 +1659,6 @@ program.addCommand(createProcess3dCommand());
 program.addCommand(createProviderCommand());
 program.addCommand(createUploadCommand());
 program.addCommand(createJobCommand());
+program.addCommand(createVoiceCommand());
 program.addCommand(createDescribeCommand());
 await program.parseAsync(process.argv);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@doufunao123/asset-gateway",
-  "version": "0.21.0",
+  "version": "0.22.0",
   "description": "Universal asset generation gateway CLI",
   "type": "module",
   "bin": {
@@ -27,7 +27,7 @@
     "node": ">=20"
   },
   "dependencies": {
-    "@doufunao123/assetforge-sdk": "^0.4.0",
+    "@doufunao123/assetforge-sdk": "^0.6.0",
     "commander": "^13.1.0"
   },
   "devDependencies": {