♻️ refactor: refactor to class usage and ui

.dumirc.ts

@@ -4,7 +4,6 @@ import path from 'node:path';
 import { description, homepage, name } from './package.json';
 
 const isProduction = process.env.NODE_ENV === 'production';
-const isWin = process.platform === 'win32';
 
 const themeConfig = {
   actions: [
@@ -42,17 +41,20 @@ const themeConfig = {
 };
 
 export default defineConfig({
+  /* eslint-disable sort-keys-fix/sort-keys-fix */
   alias: {
     '@lobehub/tts/react': path.join(__dirname, './src/react'),
   },
+  /* eslint-enable */
   apiParser: isProduction ? {} : false,
   base: '/',
   define: {
     'process.env': process.env,
   },
   favicons: ['https://npm.elemecdn.com/@lobehub/assets-favicons/assets/favicon.ico'],
   locales: [{ id: 'en-US', name: 'English' }],
-  mfsu: isWin ? undefined : {},
+  // mfsu: isWin ? undefined : {},
+  mfsu: false,
   npmClient: 'pnpm',
   publicPath: '/',
   resolve: {

.fatherrc.ts

@@ -4,6 +4,9 @@ export default defineConfig({
   esm: {
     output: 'dist',
     overrides: {
+      'src/core': {
+        output: 'core',
+      },
       'src/react': {
         output: 'react',
       },

.gitignore

@@ -55,5 +55,7 @@ next-env.d.ts
 .env
 public/*.js
 bun.lockb
-react
-server
+react/**
+server/**
+
+core/**

api/edge-speech.ts

@@ -1,5 +1,5 @@
-import { createEdgeSpeechComletion } from '../src/server/createEdgeSpeechComletion';
-import { EdgeSpeechPayload } from '../src/server/types';
+import { EdgeSpeechPayload } from '@/core';
+import { createEdgeSpeech } from '@/core/EdgeSpeechTTS/createEdgeSpeech';
 
 export const config = {
   runtime: 'edge',
@@ -10,5 +10,5 @@ export default async (req: Request) => {
 
   const payload = (await req.json()) as EdgeSpeechPayload;
 
-  return createEdgeSpeechComletion({ payload });
+  return createEdgeSpeech({ payload });
 };

api/microsoft-speech.ts

@@ -1,5 +1,5 @@
-import { createMicrosoftSpeechComletion } from '../src/server/createMicrosoftSpeechComletion';
-import { MicrosoftSpeechPayload } from '../src/server/types';
+import { MicrosoftSpeechPayload } from '@/core';
+import { createMicrosoftSpeech } from '@/core/MicrosoftSpeechTTS/createMicrosoftSpeech';
 
 export const config = {
   runtime: 'edge',
@@ -9,5 +9,5 @@ export default async (req: Request) => {
   if (req.method !== 'POST') return new Response('Method Not Allowed', { status: 405 });
   const payload = (await req.json()) as MicrosoftSpeechPayload;
 
-  return createMicrosoftSpeechComletion({ payload });
+  return createMicrosoftSpeech({ payload });
 };

api/open-stt.ts

@@ -1,7 +1,7 @@
 import OpenAI from 'openai';
 
-import { createOpenaiAudioTranscriptionsCompletion } from '../src/server/createOpenaiAudioTranscriptionsCompletion';
-import { OpenAISTTPayload } from '../src/server/types';
+import { OpenAISTTPayload } from '@/core';
+import { createOpenaiAudioTranscriptions } from '@/server/createOpenaiAudioTranscriptions';
 
 export const config = {
   runtime: 'edge',
@@ -18,7 +18,7 @@ export default async (req: Request) => {
   const payload = (await req.json()) as OpenAISTTPayload;
 
   const openai = new OpenAI({ apiKey: OPENAI_API_KEY, baseURL: OPENAI_PROXY_URL });
-  const res = await createOpenaiAudioTranscriptionsCompletion({ openai, payload });
+  const res = await createOpenaiAudioTranscriptions({ openai, payload });
 
   return new Response(JSON.stringify(res), {
     headers: {

api/openai-tts.ts

@@ -1,7 +1,7 @@
 import OpenAI from 'openai';
 
-import { createOpenaiAudioSpeechCompletion } from '../src/server/createOpenaiAudioSpeechCompletion';
-import { OpenAITTSPayload } from '../src/server/types';
+import { OpenAITTSPayload } from '@/core';
+import { createOpenaiAudioSpeech } from '@/server/createOpenaiAudioSpeech';
 
 export const config = {
   runtime: 'edge',
@@ -17,5 +17,5 @@ export default async (req: Request) => {
 
   const openai = new OpenAI({ apiKey: OPENAI_API_KEY, baseURL: OPENAI_PROXY_URL });
 
-  return createOpenaiAudioSpeechCompletion({ openai, payload });
+  return createOpenaiAudioSpeech({ openai, payload });
 };

package.json

@@ -17,9 +17,9 @@
   "exports": {
     "./package.json": "./package.json",
     ".": {
-      "types": "./dist/index.d.ts",
-      "import": "./dist/index.js",
-      "module": "./dist/index.js"
+      "types": "./core/index.d.ts",
+      "import": "./core/index.js",
+      "module": "./core/index.js"
     },
     "./server": {
       "types": "./server/index.d.ts",
@@ -32,22 +32,21 @@
       "module": "./react/index.js"
     }
   },
-  "main": "dist/index.js",
-  "module": "dist/index.js",
-  "types": "dist/index.d.ts",
+  "main": "./core/index.js",
+  "module": "./core/index.js",
+  "types": "./core/index.d.ts",
   "files": [
-    "dist",
+    "core",
     "react",
     "server"
   ],
   "scripts": {
     "build": "father build",
     "ci": "npm run lint && npm run type-check && npm run doctor",
-    "clean": "rm react.d.ts server.d.ts",
     "dev": "father dev",
     "docs:build": "npm run setup && npm run build && dumi build",
     "docs:build-analyze": "ANALYZE=1 dumi build",
-    "docs:dev": "npm run clean && npm run setup && dumi dev",
+    "docs:dev": "npm run setup && dumi dev",
     "doctor": "father doctor",
     "lint": "eslint \"{src,api,lib}/**/*.{js,jsx,ts,tsx}\" --fix",
     "lint:md": "remark . --quiet --frail --output",

src/server/createEdgeSpeechComletion.ts → src/core/EdgeSpeechTTS/createEdgeSpeech.ts

@@ -1,13 +1,26 @@
 import qs from 'query-string';
 import { v4 as uuidv4 } from 'uuid';
 
-import { EDGE_API_TOKEN, EDGE_SPEECH_URL } from '../const/api';
-import { EdgeSpeechPayload } from '../server/types';
-import { genSSML } from '../utils/genSSML';
-import { genSendContent } from '../utils/genSendContent';
-import { getHeadersAndData } from '../utils/getHeadersAndData';
+import { SsmlOptions, genSSML } from '@/core/utils/genSSML';
+import { genSendContent } from '@/core/utils/genSendContent';
+import { getHeadersAndData } from '@/core/utils/getHeadersAndData';
 
-const configConent = JSON.stringify({
+export interface EdgeSpeechPayload {
+  /**
+   * @title 语音合成的文本
+   */
+  input: string;
+  /**
+   * @title SSML 语音合成的配置
+   */
+  options: Pick<SsmlOptions, 'voice'>;
+}
+
+const EDGE_SPEECH_URL =
+  'wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1';
+const EDGE_API_TOKEN = '6A5AA1D4EAFF4E9FB37E23D68491D6F4';
+
+const configContent = JSON.stringify({
   context: {
     synthesis: {
       audio: {
@@ -37,26 +50,27 @@ const genHeader = (connectId: string) => {
   };
 };
 
-interface CreateEdgeSpeechComletionOptions {
+export interface CreateEdgeSpeechCompletionOptions {
   payload: EdgeSpeechPayload;
 }
 
-export const createEdgeSpeechComletion = async ({
-  payload,
-}: CreateEdgeSpeechComletionOptions): Promise<Response> => {
+export const createEdgeSpeech = async (
+  { payload }: CreateEdgeSpeechCompletionOptions,
+  { proxyUrl, token }: { proxyUrl?: string; token?: string } = {},
+): Promise<Response> => {
   const { input, options } = payload;
 
   const connectId = uuidv4().replaceAll('-', '');
   const url = qs.stringifyUrl({
     query: {
       ConnectionId: connectId,
-      TrustedClientToken: EDGE_API_TOKEN,
+      TrustedClientToken: token ? token : EDGE_API_TOKEN,
     },
-    url: EDGE_SPEECH_URL,
+    url: proxyUrl ? proxyUrl : EDGE_SPEECH_URL,
   });
 
   const { configHeader, contentHeader } = genHeader(connectId);
-  const config = genSendContent(configHeader, configConent);
+  const config = genSendContent(configHeader, configContent);
   const content = genSendContent(contentHeader, genSSML(input, options));
 
   return new Promise((resolve, reject) => {

src/core/EdgeSpeechTTS/index.ts

@@ -0,0 +1,57 @@
+import edgeVoiceList from '@/core/EdgeSpeechTTS/edgeVoiceList';
+import voiceName from '@/core/data/voiceList';
+import { arrayBufferConvert } from '@/core/utils/arrayBufferConvert';
+import { SsmlOptions } from '@/core/utils/genSSML';
+import { getVoiceLocaleOptions } from '@/core/utils/getVoiceList';
+
+import { createEdgeSpeech } from './createEdgeSpeech';
+import { getEdgeVoiceOptions } from './options';
+
+export interface EdgeSpeechPayload {
+  /**
+   * @title 语音合成的文本
+   */
+  input: string;
+  /**
+   * @title SSML 语音合成的配置
+   */
+  options: Pick<SsmlOptions, 'voice'>;
+}
+
+export class EdgeSpeechTTS {
+  private locale?: string;
+  private BASE_URL: string | undefined;
+
+  constructor({ baseURL, locale }: { baseURL?: string; locale?: string } = {}) {
+    this.locale = locale;
+    this.BASE_URL = baseURL;
+  }
+
+  get voiceOptions() {
+    return getEdgeVoiceOptions(this.locale);
+  }
+
+  static localeOptions = getVoiceLocaleOptions();
+  static voiceList = edgeVoiceList;
+  static voiceName = voiceName;
+  static completion = createEdgeSpeech;
+
+  private fetch = async (payload: EdgeSpeechPayload) => {
+    const response = await (this.BASE_URL
+      ? fetch(this.BASE_URL, { body: JSON.stringify(payload), method: 'POST' })
+      : createEdgeSpeech({ payload }));
+
+    if (!response.ok) {
+      throw new Error('Network response was not ok');
+    }
+    return response;
+  };
+
+  create = async (payload: EdgeSpeechPayload): Promise<AudioBuffer> => {
+    const response = await this.fetch(payload);
+
+    const arrayBuffer = await response.arrayBuffer();
+
+    return arrayBufferConvert(arrayBuffer);
+  };
+}