Merge branch 'feat/ocr' into feat/ocr-translate

2025-12-29 14:31:35 +08:00 · 2025-08-24 00:02:17 +08:00 · 2025-08-24 00:02:17 +08:00 · 07603ae9db
commit 07603ae9db
parent 7d1c775afd 32796ba03c
21 changed files with 803 additions and 137 deletions
--- a/.yarn/patches/tesseract.js-npm-6.0.1-2562a7e46d.patch
+++ b/.yarn/patches/tesseract.js-npm-6.0.1-2562a7e46d.patch
@ -0,0 +1,348 @@
+diff --git a/src/constants/languages.d.ts b/src/constants/languages.d.ts
+new file mode 100644
+index 0000000000000000000000000000000000000000..6a2ba5086187622b8ca8887bcc7406018fba8a89
+--- /dev/null
+++ b/src/constants/languages.d.ts
+@@ -0,0 +1,43 @@
+/**
+ * Languages with existing tesseract traineddata
+ * https://tesseract-ocr.github.io/tessdoc/Data-Files#data-files-for-version-400-november-29-2016
+ */
+
+// Define the language codes as string literals
+type LanguageCode =
+  | 'afr' | 'amh' | 'ara' | 'asm' | 'aze' | 'aze_cyrl' | 'bel' | 'ben' | 'bod' | 'bos'
+  | 'bul' | 'cat' | 'ceb' | 'ces' | 'chi_sim' | 'chi_tra' | 'chr' | 'cym' | 'dan' | 'deu'
+  | 'dzo' | 'ell' | 'eng' | 'enm' | 'epo' | 'est' | 'eus' | 'fas' | 'fin' | 'fra'
+  | 'frk' | 'frm' | 'gle' | 'glg' | 'grc' | 'guj' | 'hat' | 'heb' | 'hin' | 'hrv'
+  | 'hun' | 'iku' | 'ind' | 'isl' | 'ita' | 'ita_old' | 'jav' | 'jpn' | 'kan' | 'kat'
+  | 'kat_old' | 'kaz' | 'khm' | 'kir' | 'kor' | 'kur' | 'lao' | 'lat' | 'lav' | 'lit'
+  | 'mal' | 'mar' | 'mkd' | 'mlt' | 'msa' | 'mya' | 'nep' | 'nld' | 'nor' | 'ori'
+  | 'pan' | 'pol' | 'por' | 'pus' | 'ron' | 'rus' | 'san' | 'sin' | 'slk' | 'slv'
+  | 'spa' | 'spa_old' | 'sqi' | 'srp' | 'srp_latn' | 'swa' | 'swe' | 'syr' | 'tam' | 'tel'
+  | 'tgk' | 'tgl' | 'tha' | 'tir' | 'tur' | 'uig' | 'ukr' | 'urd' | 'uzb' | 'uzb_cyrl'
+  | 'vie' | 'yid';
+
+// Define the language keys as string literals
+type LanguageKey =
+  | 'AFR' | 'AMH' | 'ARA' | 'ASM' | 'AZE' | 'AZE_CYRL' | 'BEL' | 'BEN' | 'BOD' | 'BOS'
+  | 'BUL' | 'CAT' | 'CEB' | 'CES' | 'CHI_SIM' | 'CHI_TRA' | 'CHR' | 'CYM' | 'DAN' | 'DEU'
+  | 'DZO' | 'ELL' | 'ENG' | 'ENM' | 'EPO' | 'EST' | 'EUS' | 'FAS' | 'FIN' | 'FRA'
+  | 'FRK' | 'FRM' | 'GLE' | 'GLG' | 'GRC' | 'GUJ' | 'HAT' | 'HEB' | 'HIN' | 'HRV'
+  | 'HUN' | 'IKU' | 'IND' | 'ISL' | 'ITA' | 'ITA_OLD' | 'JAV' | 'JPN' | 'KAN' | 'KAT'
+  | 'KAT_OLD' | 'KAZ' | 'KHM' | 'KIR' | 'KOR' | 'KUR' | 'LAO' | 'LAT' | 'LAV' | 'LIT'
+  | 'MAL' | 'MAR' | 'MKD' | 'MLT' | 'MSA' | 'MYA' | 'NEP' | 'NLD' | 'NOR' | 'ORI'
+  | 'PAN' | 'POL' | 'POR' | 'PUS' | 'RON' | 'RUS' | 'SAN' | 'SIN' | 'SLK' | 'SLV'
+  | 'SPA' | 'SPA_OLD' | 'SQI' | 'SRP' | 'SRP_LATN' | 'SWA' | 'SWE' | 'SYR' | 'TAM' | 'TEL'
+  | 'TGK' | 'TGL' | 'THA' | 'TIR' | 'TUR' | 'UIG' | 'UKR' | 'URD' | 'UZB' | 'UZB_CYRL'
+  | 'VIE' | 'YID';
+
+// Create a mapped type to ensure each key maps to its specific value
+type LanguagesMap = {
+  [K in LanguageKey]: LanguageCode;
+};
+
+// Declare the exported constant with the specific type
+export const LANGUAGES: LanguagesMap;
+
+// Export the individual types for use in other modules
+export type { LanguageCode, LanguageKey, LanguagesMap };
+\ No newline at end of file
+diff --git a/src/index.d.ts b/src/index.d.ts
+index 1f5a9c8094fe4de7983467f9efb43bdb4de535f2..16dc95cf68663673e37e189b719cb74897b7735f 100644
+--- a/src/index.d.ts
+++ b/src/index.d.ts
+@@ -1,31 +1,74 @@
+// Import the languages types
+import { LanguagesMap } from "./constants/languages";
+
+/// <reference types="node" />
+
+ declare namespace Tesseract {
+-  function createScheduler(): Scheduler
+-  function createWorker(langs?: string | string[] | Lang[], oem?: OEM, options?: Partial<WorkerOptions>, config?: string | Partial<InitOptions>): Promise<Worker>
+-  function setLogging(logging: boolean): void
+-  function recognize(image: ImageLike, langs?: string, options?: Partial<WorkerOptions>): Promise<RecognizeResult>
+-  function detect(image: ImageLike, options?: Partial<WorkerOptions>): any
+  function createScheduler(): Scheduler;
+  function createWorker(
+    langs?: LanguageCode | LanguageCode[] | Lang[],
+    oem?: OEM,
+    options?: Partial<WorkerOptions>,
+    config?: string | Partial<InitOptions>
+  ): Promise<Worker>;
+  function setLogging(logging: boolean): void;
+  function recognize(
+    image: ImageLike,
+    langs?: LanguageCode,
+    options?: Partial<WorkerOptions>
+  ): Promise<RecognizeResult>;
+  function detect(image: ImageLike, options?: Partial<WorkerOptions>): any;
+
+  // Export languages constant
+  const languages: LanguagesMap;
+
+  type LanguageCode = import("./constants/languages").LanguageCode;
+  type LanguageKey = import("./constants/languages").LanguageKey;
+ 
+   interface Scheduler {
+-    addWorker(worker: Worker): string
+-    addJob(action: 'recognize', ...args: Parameters<Worker['recognize']>): Promise<RecognizeResult>
+-    addJob(action: 'detect', ...args: Parameters<Worker['detect']>): Promise<DetectResult>
+-    terminate(): Promise<any>
+-    getQueueLen(): number
+-    getNumWorkers(): number
+    addWorker(worker: Worker): string;
+    addJob(
+      action: "recognize",
+      ...args: Parameters<Worker["recognize"]>
+    ): Promise<RecognizeResult>;
+    addJob(
+      action: "detect",
+      ...args: Parameters<Worker["detect"]>
+    ): Promise<DetectResult>;
+    terminate(): Promise<any>;
+    getQueueLen(): number;
+    getNumWorkers(): number;
+   }
+ 
+   interface Worker {
+-    load(jobId?: string): Promise<ConfigResult>
+-    writeText(path: string, text: string, jobId?: string): Promise<ConfigResult>
+-    readText(path: string, jobId?: string): Promise<ConfigResult>
+-    removeText(path: string, jobId?: string): Promise<ConfigResult>
+-    FS(method: string, args: any[], jobId?: string): Promise<ConfigResult>
+-    reinitialize(langs?: string | Lang[], oem?: OEM, config?: string | Partial<InitOptions>, jobId?: string): Promise<ConfigResult>
+-    setParameters(params: Partial<WorkerParams>, jobId?: string): Promise<ConfigResult>
+-    getImage(type: imageType): string
+-    recognize(image: ImageLike, options?: Partial<RecognizeOptions>, output?: Partial<OutputFormats>, jobId?: string): Promise<RecognizeResult>
+-    detect(image: ImageLike, jobId?: string): Promise<DetectResult>
+-    terminate(jobId?: string): Promise<ConfigResult>
+    load(jobId?: string): Promise<ConfigResult>;
+    writeText(
+      path: string,
+      text: string,
+      jobId?: string
+    ): Promise<ConfigResult>;
+    readText(path: string, jobId?: string): Promise<ConfigResult>;
+    removeText(path: string, jobId?: string): Promise<ConfigResult>;
+    FS(method: string, args: any[], jobId?: string): Promise<ConfigResult>;
+    reinitialize(
+      langs?: string | Lang[],
+      oem?: OEM,
+      config?: string | Partial<InitOptions>,
+      jobId?: string
+    ): Promise<ConfigResult>;
+    setParameters(
+      params: Partial<WorkerParams>,
+      jobId?: string
+    ): Promise<ConfigResult>;
+    getImage(type: imageType): string;
+    recognize(
+      image: ImageLike,
+      options?: Partial<RecognizeOptions>,
+      output?: Partial<OutputFormats>,
+      jobId?: string
+    ): Promise<RecognizeResult>;
+    detect(image: ImageLike, jobId?: string): Promise<DetectResult>;
+    terminate(jobId?: string): Promise<ConfigResult>;
+   }
+ 
+   interface Lang {
+@@ -34,43 +77,43 @@ declare namespace Tesseract {
+   }
+ 
+   interface InitOptions {
+-    load_system_dawg: string
+-    load_freq_dawg: string
+-    load_unambig_dawg: string
+-    load_punc_dawg: string
+-    load_number_dawg: string
+-    load_bigram_dawg: string
+-  }
+-  
+-  type LoggerMessage = {
+-    jobId: string
+-    progress: number
+-    status: string
+-    userJobId: string
+-    workerId: string
+    load_system_dawg: string;
+    load_freq_dawg: string;
+    load_unambig_dawg: string;
+    load_punc_dawg: string;
+    load_number_dawg: string;
+    load_bigram_dawg: string;
+   }
+-  
+
+  type LoggerMessage = {
+    jobId: string;
+    progress: number;
+    status: string;
+    userJobId: string;
+    workerId: string;
+  };
+
+   interface WorkerOptions {
+-    corePath: string
+-    langPath: string
+-    cachePath: string
+-    dataPath: string
+-    workerPath: string
+-    cacheMethod: string
+-    workerBlobURL: boolean
+-    gzip: boolean
+-    legacyLang: boolean
+-    legacyCore: boolean
+-    logger: (arg: LoggerMessage) => void,
+-    errorHandler: (arg: any) => void
+    corePath: string;
+    langPath: string;
+    cachePath: string;
+    dataPath: string;
+    workerPath: string;
+    cacheMethod: string;
+    workerBlobURL: boolean;
+    gzip: boolean;
+    legacyLang: boolean;
+    legacyCore: boolean;
+    logger: (arg: LoggerMessage) => void;
+    errorHandler: (arg: any) => void;
+   }
+   interface WorkerParams {
+-    tessedit_pageseg_mode: PSM
+-    tessedit_char_whitelist: string
+-    tessedit_char_blacklist: string
+-    preserve_interword_spaces: string
+-    user_defined_dpi: string
+-    [propName: string]: any
+    tessedit_pageseg_mode: PSM;
+    tessedit_char_whitelist: string;
+    tessedit_char_blacklist: string;
+    preserve_interword_spaces: string;
+    user_defined_dpi: string;
+    [propName: string]: any;
+   }
+   interface OutputFormats {
+     text: boolean;
+@@ -88,36 +131,36 @@ declare namespace Tesseract {
+     debug: boolean;
+   }
+   interface RecognizeOptions {
+-    rectangle: Rectangle
+-    pdfTitle: string
+-    pdfTextOnly: boolean
+-    rotateAuto: boolean
+-    rotateRadians: number
+    rectangle: Rectangle;
+    pdfTitle: string;
+    pdfTextOnly: boolean;
+    rotateAuto: boolean;
+    rotateRadians: number;
+   }
+   interface ConfigResult {
+-    jobId: string
+-    data: any
+    jobId: string;
+    data: any;
+   }
+   interface RecognizeResult {
+-    jobId: string
+-    data: Page
+    jobId: string;
+    data: Page;
+   }
+   interface DetectResult {
+-    jobId: string
+-    data: DetectData
+    jobId: string;
+    data: DetectData;
+   }
+   interface DetectData {
+-    tesseract_script_id: number | null
+-    script: string | null
+-    script_confidence: number | null
+-    orientation_degrees: number | null
+-    orientation_confidence: number | null
+    tesseract_script_id: number | null;
+    script: string | null;
+    script_confidence: number | null;
+    orientation_degrees: number | null;
+    orientation_confidence: number | null;
+   }
+   interface Rectangle {
+-    left: number
+-    top: number
+-    width: number
+-    height: number
+    left: number;
+    top: number;
+    width: number;
+    height: number;
+   }
+   enum OEM {
+     TESSERACT_ONLY,
+@@ -126,28 +169,36 @@ declare namespace Tesseract {
+     DEFAULT,
+   }
+   enum PSM {
+-    OSD_ONLY = '0',
+-    AUTO_OSD = '1',
+-    AUTO_ONLY = '2',
+-    AUTO = '3',
+-    SINGLE_COLUMN = '4',
+-    SINGLE_BLOCK_VERT_TEXT = '5',
+-    SINGLE_BLOCK = '6',
+-    SINGLE_LINE = '7',
+-    SINGLE_WORD = '8',
+-    CIRCLE_WORD = '9',
+-    SINGLE_CHAR = '10',
+-    SPARSE_TEXT = '11',
+-    SPARSE_TEXT_OSD = '12',
+-    RAW_LINE = '13'
+    OSD_ONLY = "0",
+    AUTO_OSD = "1",
+    AUTO_ONLY = "2",
+    AUTO = "3",
+    SINGLE_COLUMN = "4",
+    SINGLE_BLOCK_VERT_TEXT = "5",
+    SINGLE_BLOCK = "6",
+    SINGLE_LINE = "7",
+    SINGLE_WORD = "8",
+    CIRCLE_WORD = "9",
+    SINGLE_CHAR = "10",
+    SPARSE_TEXT = "11",
+    SPARSE_TEXT_OSD = "12",
+    RAW_LINE = "13",
+   }
+   const enum imageType {
+     COLOR = 0,
+     GREY = 1,
+-    BINARY = 2
+    BINARY = 2,
+   }
+-  type ImageLike = string | HTMLImageElement | HTMLCanvasElement | HTMLVideoElement
+-    | CanvasRenderingContext2D | File | Blob | Buffer | OffscreenCanvas;
+  type ImageLike =
+    | string
+    | HTMLImageElement
+    | HTMLCanvasElement
+    | HTMLVideoElement
+    | CanvasRenderingContext2D
+    | File
+    | Blob
+    | (typeof Buffer extends undefined ? never : Buffer)
+    | OffscreenCanvas;
+   interface Block {
+     paragraphs: Paragraph[];
+     text: string;
+@@ -179,7 +230,7 @@ declare namespace Tesseract {
+     text: string;
+     confidence: number;
+     baseline: Baseline;
+-    rowAttributes: RowAttributes
+    rowAttributes: RowAttributes;
+     bbox: Bbox;
+   }
+   interface Paragraph {
--- a/package.json
+++ b/package.json
@ -79,7 +79,7 @@
    "officeparser": "^4.2.0",
    "os-proxy-config": "^1.1.2",
    "selection-hook": "^1.0.11",
-    "tesseract.js": "^6.0.1",
+    "tesseract.js": "patch:tesseract.js@npm%3A6.0.1#~/.yarn/patches/tesseract.js-npm-6.0.1-2562a7e46d.patch",
    "turndown": "7.2.0"
  },
  "devDependencies": {
@ -293,7 +293,8 @@
    "pdf-parse@npm:1.1.1": "patch:pdf-parse@npm%3A1.1.1#~/.yarn/patches/pdf-parse-npm-1.1.1-04a6109b2a.patch",
    "pkce-challenge@npm:^4.1.0": "patch:pkce-challenge@npm%3A4.1.0#~/.yarn/patches/pkce-challenge-npm-4.1.0-fbc51695a3.patch",
    "undici": "6.21.2",
-    "vite": "npm:rolldown-vite@latest"
+    "vite": "npm:rolldown-vite@latest",
+    "tesseract.js@npm:*": "patch:tesseract.js@npm%3A6.0.1#~/.yarn/patches/tesseract.js-npm-6.0.1-2562a7e46d.patch"
  },
  "packageManager": "yarn@4.9.1",
  "lint-staged": {
--- a/src/main/services/ocr/tesseract/TesseractService.ts
+++ b/src/main/services/ocr/tesseract/TesseractService.ts
@ -5,118 +5,13 @@ import { ImageFileMetadata, isImageFile, OcrResult, SupportedOcrFile } from '@ty
 import { app } from 'electron'
 import fs from 'fs'
 import path from 'path'
-import Tesseract, { createWorker } from 'tesseract.js'
+import Tesseract, { createWorker, LanguageCode } from 'tesseract.js'

 const logger = loggerService.withContext('TesseractService')

-// const languageCodeMap: Record<string, string> = {
-//   'af-za': 'afr',
-//   'am-et': 'amh',
-//   'ar-sa': 'ara',
-//   'as-in': 'asm',
-//   'az-az': 'aze',
-//   'az-cyrl-az': 'aze_cyrl',
-//   'be-by': 'bel',
-//   'bn-bd': 'ben',
-//   'bo-cn': 'bod',
-//   'bs-ba': 'bos',
-//   'bg-bg': 'bul',
-//   'ca-es': 'cat',
-//   'ceb-ph': 'ceb',
-//   'cs-cz': 'ces',
-//   'zh-cn': 'chi_sim',
-//   'zh-tw': 'chi_tra',
-//   'chr-us': 'chr',
-//   'cy-gb': 'cym',
-//   'da-dk': 'dan',
-//   'de-de': 'deu',
-//   'dz-bt': 'dzo',
-//   'el-gr': 'ell',
-//   'en-us': 'eng',
-//   'enm-gb': 'enm',
-//   'eo-world': 'epo',
-//   'et-ee': 'est',
-//   'eu-es': 'eus',
-//   'fa-ir': 'fas',
-//   'fi-fi': 'fin',
-//   'fr-fr': 'fra',
-//   'frk-de': 'frk',
-//   'frm-fr': 'frm',
-//   'ga-ie': 'gle',
-//   'gl-es': 'glg',
-//   'grc-gr': 'grc',
-//   'gu-in': 'guj',
-//   'ht-ht': 'hat',
-//   'he-il': 'heb',
-//   'hi-in': 'hin',
-//   'hr-hr': 'hrv',
-//   'hu-hu': 'hun',
-//   'iu-ca': 'iku',
-//   'id-id': 'ind',
-//   'is-is': 'isl',
-//   'it-it': 'ita',
-//   'ita-it': 'ita_old',
-//   'jv-id': 'jav',
-//   'ja-jp': 'jpn',
-//   'kn-in': 'kan',
-//   'ka-ge': 'kat',
-//   'kat-ge': 'kat_old',
-//   'kk-kz': 'kaz',
-//   'km-kh': 'khm',
-//   'ky-kg': 'kir',
-//   'ko-kr': 'kor',
-//   'ku-tr': 'kur',
-//   'la-la': 'lao',
-//   'la-va': 'lat',
-//   'lv-lv': 'lav',
-//   'lt-lt': 'lit',
-//   'ml-in': 'mal',
-//   'mr-in': 'mar',
-//   'mk-mk': 'mkd',
-//   'mt-mt': 'mlt',
-//   'ms-my': 'msa',
-//   'my-mm': 'mya',
-//   'ne-np': 'nep',
-//   'nl-nl': 'nld',
-//   'no-no': 'nor',
-//   'or-in': 'ori',
-//   'pa-in': 'pan',
-//   'pl-pl': 'pol',
-//   'pt-pt': 'por',
-//   'ps-af': 'pus',
-//   'ro-ro': 'ron',
-//   'ru-ru': 'rus',
-//   'sa-in': 'san',
-//   'si-lk': 'sin',
-//   'sk-sk': 'slk',
-//   'sl-si': 'slv',
-//   'es-es': 'spa',
-//   'spa-es': 'spa_old',
-//   'sq-al': 'sqi',
-//   'sr-rs': 'srp',
-//   'sr-latn-rs': 'srp_latn',
-//   'sw-tz': 'swa',
-//   'sv-se': 'swe',
-//   'syr-sy': 'syr',
-//   'ta-in': 'tam',
-//   'te-in': 'tel',
-//   'tg-tj': 'tgk',
-//   'tl-ph': 'tgl',
-//   'th-th': 'tha',
-//   'ti-er': 'tir',
-//   'tr-tr': 'tur',
-//   'ug-cn': 'uig',
-//   'uk-ua': 'ukr',
-//   'ur-pk': 'urd',
-//   'uz-uz': 'uzb',
-//   'uz-cyrl-uz': 'uzb_cyrl',
-//   'vi-vn': 'vie',
-//   'yi-us': 'yid'
-// }
-
 // config
 const MB_SIZE_THRESHOLD = 50
-const tesseractLangs = ['chi_sim', 'chi_tra', 'eng']
+const tesseractLangs = ['chi_sim', 'chi_tra', 'eng'] satisfies LanguageCode[]
 enum TesseractLangsDownloadUrl {
  CN = 'https://gitcode.com/beyondkmp/tessdata/releases/download/4.1.0/',
  GLOBAL = 'https://github.com/tesseract-ocr/tessdata/raw/main/'
--- a/src/renderer/src/config/ocr.ts
+++ b/src/renderer/src/config/ocr.ts
@ -1,14 +1,31 @@
-import { BuiltinOcrProvider, ImageOcrProvider, OcrProviderCapability } from '@renderer/types'
+import {
+  BuiltinOcrProvider,
+  BuiltinOcrProviderId,
+  ImageOcrProvider,
+  OcrProviderCapability,
+  OcrTesseractProvider
+} from '@renderer/types'

-const tesseract: BuiltinOcrProvider & ImageOcrProvider = {
+const tesseract: BuiltinOcrProvider & ImageOcrProvider & OcrTesseractProvider = {
  id: 'tesseract',
  name: 'Tesseract',
  capabilities: {
    image: true
+  },
+  config: {
+    langs: {
+      chi_sim: true,
+      chi_tra: true,
+      eng: true
+    }
  }
-} as const
+} as const satisfies OcrTesseractProvider

-export const BUILTIN_OCR_PROVIDERS: BuiltinOcrProvider[] = [tesseract] as const
+export const BUILTIN_OCR_PROVIDERS_MAP = {
+  tesseract
+} as const satisfies Record<BuiltinOcrProviderId, BuiltinOcrProvider>
+
+export const BUILTIN_OCR_PROVIDERS: BuiltinOcrProvider[] = Object.values(BUILTIN_OCR_PROVIDERS_MAP)

 export const DEFAULT_OCR_PROVIDER = {
  image: tesseract
--- a/src/renderer/src/hooks/useOcrProvider.ts
+++ b/src/renderer/src/hooks/useOcrProvider.ts
@ -0,0 +1,84 @@
+import { loggerService } from '@logger'
+import { BUILTIN_OCR_PROVIDERS_MAP } from '@renderer/config/ocr'
+import { useAppSelector } from '@renderer/store'
+import { addOcrProvider, removeOcrProvider, updateOcrProviderConfig } from '@renderer/store/ocr'
+import { isBuiltinOcrProviderId, OcrProvider, OcrProviderConfig } from '@renderer/types'
+import { useTranslation } from 'react-i18next'
+import { useDispatch } from 'react-redux'
+
+const logger = loggerService.withContext('useOcrProvider')
+
+export const useOcrProviders = () => {
+  const providers = useAppSelector((state) => state.ocr.providers)
+  const dispatch = useDispatch()
+  const { t } = useTranslation()
+
+  /**
+   * 添加一个新的OCR服务提供者
+   * @param provider - OCR提供者对象，包含id和其他配置信息
+   * @throws {Error} 当尝试添加一个已存在ID的提供者时抛出错误
+   */
+  const addProvider = (provider: OcrProvider) => {
+    if (providers.some((p) => p.id === provider.id)) {
+      const msg = `Provider with id ${provider.id} already exists`
+      logger.error(msg)
+      window.message.error(t('ocr.error.provider.existing'))
+      throw new Error(msg)
+    }
+    dispatch(addOcrProvider(provider))
+  }
+
+  /**
+   * 移除一个OCR服务提供者
+   * @param id - 要移除的OCR提供者ID
+   * @throws {Error} 当尝试移除一个内置提供商时抛出错误
+   */
+  const removeProvider = (id: string) => {
+    if (isBuiltinOcrProviderId(id)) {
+      const msg = `Cannot remove builtin provider ${id}`
+      logger.error(msg)
+      window.message.error(t('ocr.error.provider.cannot_remove_builtin'))
+      throw new Error(msg)
+    }
+
+    dispatch(removeOcrProvider(id))
+  }
+
+  return { providers, addProvider, removeProvider }
+}
+
+export const useOcrProvider = (id: string) => {
+  const { t } = useTranslation()
+  const dispatch = useDispatch()
+  const { providers, addProvider } = useOcrProviders()
+  let provider = providers.find((p) => p.id === id)
+
+  // safely fallback
+  if (!provider) {
+    logger.error(`Ocr Provider ${id} not found`)
+    window.message.error(t('ocr.error.provider.not_found'))
+    if (isBuiltinOcrProviderId(id)) {
+      try {
+        addProvider(BUILTIN_OCR_PROVIDERS_MAP[id])
+      } catch (e) {
+        logger.warn(`Add ${BUILTIN_OCR_PROVIDERS_MAP[id].name} failed. Just use temp provider from config.`)
+        window.message.warning(t('ocr.warning.provider.fallback', { name: BUILTIN_OCR_PROVIDERS_MAP[id].name }))
+      } finally {
+        provider = BUILTIN_OCR_PROVIDERS_MAP[id]
+      }
+    } else {
+      logger.warn(`Fallback to tesseract`)
+      window.message.warning(t('ocr.warning.provider.fallback', { name: 'Tesseract' }))
+      provider = BUILTIN_OCR_PROVIDERS_MAP.tesseract
+    }
+  }
+
+  const updateConfig = (update: Partial<OcrProviderConfig>) => {
+    dispatch(updateOcrProviderConfig({ id: provider.id, update }))
+  }
+
+  return {
+    provider,
+    updateConfig
+  }
+}
--- a/src/renderer/src/i18n/locales/en-us.json
+++ b/src/renderer/src/i18n/locales/en-us.json
@ -1566,12 +1566,23 @@
  },
  "ocr": {
    "error": {
+      "provider": {
+        "cannot_remove_builtin": "Cannot delete built-in provider",
+        "existing": "The provider already exists",
+        "not_found": "OCR provider does not exist",
+        "update_failed": "Failed to update configuration"
+      },
      "unknown": "An error occurred during the OCR process"
    },
    "file": {
      "not_supported": "Unsupported file type {{type}}"
    },
-    "processing": "OCR processing..."
+    "processing": "OCR processing...",
+    "warning": {
+      "provider": {
+        "fallback": "Reverted to {{name}}, which may cause issues"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3498,6 +3509,10 @@
          "error": {
            "provider_not_found": "The provider does not exist"
          },
+          "tesseract": {
+            "langs": "Supported languages",
+            "temp_tooltip": "Currently only Chinese and English are supported"
+          },
          "title": "Image"
        },
        "image_provider": "OCR service provider",
--- a/src/renderer/src/i18n/locales/ja-jp.json
+++ b/src/renderer/src/i18n/locales/ja-jp.json
@ -1566,12 +1566,23 @@
  },
  "ocr": {
    "error": {
+      "provider": {
+        "cannot_remove_builtin": "組み込みプロバイダーは削除できません",
+        "existing": "プロバイダーはすでに存在します",
+        "not_found": "OCRプロバイダーが存在しません",
+        "update_failed": "更新構成に失敗しました"
+      },
      "unknown": "OCR処理中にエラーが発生しました"
    },
    "file": {
      "not_supported": "サポートされていないファイルタイプ {{type}}"
    },
-    "processing": "OCR処理中..."
+    "processing": "OCR処理中...",
+    "warning": {
+      "provider": {
+        "fallback": "{{name}} に戻されました。これにより問題が発生する可能性があります。"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3498,6 +3509,10 @@
          "error": {
            "provider_not_found": "該提供者は存在しません"
          },
+          "tesseract": {
+            "langs": "サポートされている言語",
+            "temp_tooltip": "現在のところ、中国語と英語のみをサポートしています"
+          },
          "title": "画像"
        },
        "image_provider": "OCRサービスプロバイダー",
--- a/src/renderer/src/i18n/locales/ru-ru.json
+++ b/src/renderer/src/i18n/locales/ru-ru.json
@ -1566,12 +1566,23 @@
  },
  "ocr": {
    "error": {
+      "provider": {
+        "cannot_remove_builtin": "Не удается удалить встроенного поставщика",
+        "existing": "Поставщик уже существует",
+        "not_found": "Поставщик OCR отсутствует",
+        "update_failed": "Обновление конфигурации не удалось"
+      },
      "unknown": "Произошла ошибка в процессе распознавания текста"
    },
    "file": {
      "not_supported": "Неподдерживаемый тип файла {{type}}"
    },
-    "processing": "Обработка OCR..."
+    "processing": "Обработка OCR...",
+    "warning": {
+      "provider": {
+        "fallback": "Возвращено к {{name}}, это может вызвать проблемы"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3498,6 +3509,10 @@
          "error": {
            "provider_not_found": "Поставщик не существует"
          },
+          "tesseract": {
+            "langs": "Поддерживаемые языки",
+            "temp_tooltip": "На данный момент поддерживаются только китайский и английский языки"
+          },
          "title": "Изображение"
        },
        "image_provider": "Поставщик услуг OCR",
--- a/src/renderer/src/i18n/locales/zh-cn.json
+++ b/src/renderer/src/i18n/locales/zh-cn.json
@ -1566,12 +1566,23 @@
  },
  "ocr": {
    "error": {
-      "unknown": "OCR过程发生错误"
+      "provider": {
+        "cannot_remove_builtin": "不能删除内置提供商",
+        "existing": "提供商已存在",
+        "not_found": "OCR 提供商不存在",
+        "update_failed": "更新配置失败"
+      },
+      "unknown": "OCR 过程发生错误"
    },
    "file": {
      "not_supported": "不支持的文件类型 {{type}}"
    },
-    "processing": "OCR 处理中..."
+    "processing": "OCR 处理中...",
+    "warning": {
+      "provider": {
+        "fallback": "已回退到 {{name}}，这可能导致问题"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3498,6 +3509,10 @@
          "error": {
            "provider_not_found": "该提供商不存在"
          },
+          "tesseract": {
+            "langs": "支持的语言",
+            "temp_tooltip": "目前暂时只支持中文和英文"
+          },
          "title": "图片"
        },
        "image_provider": "OCR 服务提供商",
--- a/src/renderer/src/i18n/locales/zh-tw.json
+++ b/src/renderer/src/i18n/locales/zh-tw.json
@ -1566,12 +1566,23 @@
  },
  "ocr": {
    "error": {
+      "provider": {
+        "cannot_remove_builtin": "不能刪除內建提供者",
+        "existing": "提供商已存在",
+        "not_found": "OCR 提供商不存在",
+        "update_failed": "更新配置失敗"
+      },
      "unknown": "OCR過程發生錯誤"
    },
    "file": {
      "not_supported": "不支持的文件類型 {{type}}"
    },
-    "processing": "OCR 處理中..."
+    "processing": "OCR 處理中...",
+    "warning": {
+      "provider": {
+        "fallback": "已回退到 {{name}}，這可能導致問題"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3498,6 +3509,10 @@
          "error": {
            "provider_not_found": "該提供商不存在"
          },
+          "tesseract": {
+            "langs": "支援的語言",
+            "temp_tooltip": "目前暫時只支援中文和英文"
+          },
          "title": "圖片"
        },
        "image_provider": "OCR 服務提供商",
--- a/src/renderer/src/i18n/translate/el-gr.json
+++ b/src/renderer/src/i18n/translate/el-gr.json
@ -888,6 +888,9 @@
  },
  "history": {
    "continue_chat": "Συνεχίστε το συνομιλημένο",
+    "error": {
+      "topic_not_found": "Το θέμα δεν υπάρχει"
+    },
    "locate": {
      "message": "Εφαρμογή στο μήνυμα"
    },
@ -1563,12 +1566,23 @@
  },
  "ocr": {
    "error": {
+      "provider": {
+        "cannot_remove_builtin": "Δεν είναι δυνατή η διαγραφή του ενσωματωμένου παρόχου",
+        "existing": "Ο πάροχος υπηρεσιών υπάρχει ήδη",
+        "not_found": "Ο πάροχος OCR δεν υπάρχει",
+        "update_failed": "Αποτυχία ενημέρωσης της διαμόρφωσης"
+      },
      "unknown": "Η διαδικασία OCR εμφάνισε σφάλμα"
    },
    "file": {
      "not_supported": "Μη υποστηριζόμενος τύπος αρχείου {{type}}"
    },
-    "processing": "Η επεξεργασία OCR βρίσκεται σε εξέλιξη..."
+    "processing": "Η επεξεργασία OCR βρίσκεται σε εξέλιξη...",
+    "warning": {
+      "provider": {
+        "fallback": "Επαναφέρθηκε στο {{name}}, το οποίο μπορεί να προκαλέσει προβλήματα"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3495,6 +3509,10 @@
          "error": {
            "provider_not_found": "Ο πάροχος δεν υπάρχει"
          },
+          "tesseract": {
+            "langs": "Υποστηριζόμενες γλώσσες",
+            "temp_tooltip": "Προς το παρόν υποστηρίζονται μόνο η κινεζική και η αγγλική γλώσσα"
+          },
          "title": "Εικόνα"
        },
        "image_provider": "Πάροχοι υπηρεσιών OCR",
--- a/src/renderer/src/i18n/translate/es-es.json
+++ b/src/renderer/src/i18n/translate/es-es.json
@ -888,6 +888,9 @@
  },
  "history": {
    "continue_chat": "Continuar chat",
+    "error": {
+      "topic_not_found": "El tema no existe"
+    },
    "locate": {
      "message": "Localizar mensaje"
    },
@ -1563,12 +1566,23 @@
  },
  "ocr": {
    "error": {
+      "provider": {
+        "cannot_remove_builtin": "No se puede eliminar el proveedor integrado",
+        "existing": "El proveedor ya existe",
+        "not_found": "El proveedor de OCR no existe",
+        "update_failed": "Actualización de la configuración fallida"
+      },
      "unknown": "El proceso OCR ha fallado"
    },
    "file": {
      "not_supported": "Tipo de archivo no compatible {{type}}"
    },
-    "processing": "Procesando OCR..."
+    "processing": "Procesando OCR...",
+    "warning": {
+      "provider": {
+        "fallback": "Se ha revertido a {{name}}, lo que podría causar problemas"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3495,6 +3509,10 @@
          "error": {
            "provider_not_found": "El proveedor no existe"
          },
+          "tesseract": {
+            "langs": "Idiomas compatibles",
+            "temp_tooltip": "Actualmente solo se admiten chino e inglés."
+          },
          "title": "Imagen"
        },
        "image_provider": "Proveedor de servicios OCR",
--- a/src/renderer/src/i18n/translate/fr-fr.json
+++ b/src/renderer/src/i18n/translate/fr-fr.json
@ -888,6 +888,9 @@
  },
  "history": {
    "continue_chat": "Continuer la conversation",
+    "error": {
+      "topic_not_found": "Le sujet n'existe pas"
+    },
    "locate": {
      "message": "Localiser le message"
    },
@ -1563,12 +1566,23 @@
  },
  "ocr": {
    "error": {
+      "provider": {
+        "cannot_remove_builtin": "Impossible de supprimer le fournisseur intégré",
+        "existing": "Le fournisseur existe déjà",
+        "not_found": "Le fournisseur OCR n'existe pas",
+        "update_failed": "Échec de la mise à jour de la configuration"
+      },
      "unknown": "Une erreur s'est produite lors du processus OCR"
    },
    "file": {
      "not_supported": "Type de fichier non pris en charge {{type}}"
    },
-    "processing": "Traitement OCR en cours..."
+    "processing": "Traitement OCR en cours...",
+    "warning": {
+      "provider": {
+        "fallback": "Revenu à {{name}}, ce qui pourrait entraîner des problèmes"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3495,6 +3509,10 @@
          "error": {
            "provider_not_found": "Ce fournisseur n'existe pas"
          },
+          "tesseract": {
+            "langs": "Langues prises en charge",
+            "temp_tooltip": "Pour le moment, seuls le chinois et l'anglais sont pris en charge."
+          },
          "title": "Image"
        },
        "image_provider": "Fournisseur de service OCR",
--- a/src/renderer/src/i18n/translate/pt-pt.json
+++ b/src/renderer/src/i18n/translate/pt-pt.json
@ -888,6 +888,9 @@
  },
  "history": {
    "continue_chat": "Continuar conversando",
+    "error": {
+      "topic_not_found": "Tópico inexistente"
+    },
    "locate": {
      "message": "Localizar mensagem"
    },
@ -1563,12 +1566,23 @@
  },
  "ocr": {
    "error": {
+      "provider": {
+        "cannot_remove_builtin": "Não é possível excluir o provedor integrado",
+        "existing": "O provedor já existe",
+        "not_found": "O provedor OCR não existe",
+        "update_failed": "Falha ao atualizar a configuração"
+      },
      "unknown": "O processo OCR apresentou um erro"
    },
    "file": {
      "not_supported": "Tipo de arquivo não suportado {{type}}"
    },
-    "processing": "Processamento OCR em andamento..."
+    "processing": "Processamento OCR em andamento...",
+    "warning": {
+      "provider": {
+        "fallback": "Revertido para {{name}}, o que pode causar problemas"
+      }
+    }
  },
  "ollama": {
    "keep_alive_time": {
@ -3495,6 +3509,10 @@
          "error": {
            "provider_not_found": "O provedor não existe"
          },
+          "tesseract": {
+            "langs": "Idiomas suportados",
+            "temp_tooltip": "No momento, apenas chinês e inglês são suportados."
+          },
          "title": "Imagem"
        },
        "image_provider": "Provedor de serviços OCR",
--- a/src/renderer/src/pages/settings/DocProcessSettings/OcrImageProviderSettings.tsx
+++ b/src/renderer/src/pages/settings/DocProcessSettings/OcrImageProviderSettings.tsx
@ -1,22 +1,32 @@
 import { loggerService } from '@logger'
 import { useAppSelector } from '@renderer/store'
 import { setImageOcrProvider } from '@renderer/store/ocr'
-import { isImageOcrProvider } from '@renderer/types'
+import { isImageOcrProvider, OcrProvider } from '@renderer/types'
 import { Select } from 'antd'
+import { useEffect } from 'react'
 import { useTranslation } from 'react-i18next'
 import { useDispatch } from 'react-redux'

 import { SettingRow, SettingRowTitle } from '..'

-const logger = loggerService.withContext('OcrImageProviderSettings')
+const logger = loggerService.withContext('OcrImageSettings')

-const OcrImageProviderSettings = () => {
+type Props = {
+  setProvider: (provider: OcrProvider) => void
+}
+
+const OcrImageSettings = ({ setProvider }: Props) => {
  const { t } = useTranslation()
  const providers = useAppSelector((state) => state.ocr.providers)
  const imageProvider = useAppSelector((state) => state.ocr.imageProvider)
  const imageProviders = providers.filter((p) => isImageOcrProvider(p))
  const dispatch = useDispatch()

+  // 挂载时更新外部状态
+  useEffect(() => {
+    setProvider(imageProvider)
+  }, [imageProvider, setProvider])
+
  const updateImageProvider = (id: string) => {
    const provider = imageProviders.find((p) => p.id === id)
    if (!provider) {
@ -25,6 +35,7 @@ const OcrImageProviderSettings = () => {
      return
    }

+    setProvider(provider)
    dispatch(setImageOcrProvider(provider))
  }

@ -48,4 +59,4 @@ const OcrImageProviderSettings = () => {
  )
 }

-export default OcrImageProviderSettings
+export default OcrImageSettings
--- a/src/renderer/src/pages/settings/DocProcessSettings/OcrProviderSettings.tsx
+++ b/src/renderer/src/pages/settings/DocProcessSettings/OcrProviderSettings.tsx
@ -0,0 +1,51 @@
+// import { loggerService } from '@logger'
+import { isBuiltinOcrProvider, OcrProvider } from '@renderer/types'
+import { getOcrProviderLogo } from '@renderer/utils/ocr'
+import { Avatar, Divider, Flex } from 'antd'
+import styled from 'styled-components'
+
+import { SettingTitle } from '..'
+import { OcrTesseractSettings } from './OcrTesseractSettings'
+
+// const logger = loggerService.withContext('OcrTesseractSettings')
+
+type Props = {
+  provider: OcrProvider
+}
+
+const OcrProviderSettings = ({ provider }: Props) => {
+  // const { t } = useTranslation()
+  const getProviderSettings = () => {
+    if (isBuiltinOcrProvider(provider)) {
+      switch (provider.id) {
+        case 'tesseract':
+          return <OcrTesseractSettings />
+      }
+    } else {
+      throw new Error('Not supported OCR provider')
+    }
+  }
+
+  return (
+    <>
+      <SettingTitle>
+        <Flex align="center" gap={8}>
+          <ProviderLogo shape="square" src={getOcrProviderLogo(provider.id)} size={16} />
+          <ProviderName> {provider.name}</ProviderName>
+        </Flex>
+      </SettingTitle>
+      <Divider style={{ width: '100%', margin: '10px 0' }} />
+      {getProviderSettings()}
+    </>
+  )
+}
+
+const ProviderName = styled.span`
+  font-size: 14px;
+  font-weight: 500;
+`
+const ProviderLogo = styled(Avatar)`
+  border: 0.5px solid var(--color-border);
+`
+
+export default OcrProviderSettings
--- a/src/renderer/src/pages/settings/DocProcessSettings/OcrSettings.tsx
+++ b/src/renderer/src/pages/settings/DocProcessSettings/OcrSettings.tsx
@ -1,22 +1,27 @@
 import { PictureOutlined } from '@ant-design/icons'
 import { useTheme } from '@renderer/context/ThemeProvider'
+import { useAppSelector } from '@renderer/store'
+import { OcrProvider } from '@renderer/types'
 import { Tabs, TabsProps } from 'antd'
-import { FC } from 'react'
+import { FC, useState } from 'react'
 import { useTranslation } from 'react-i18next'

 import { SettingDivider, SettingGroup, SettingTitle } from '..'
-import OcrImageProviderSettings from './OcrImageProviderSettings'
+import OcrImageSettings from './OcrImageSettings'
+import OcrProviderSettings from './OcrProviderSettings'

 const OcrSettings: FC = () => {
  const { t } = useTranslation()
  const { theme: themeMode } = useTheme()
+  const imageProvider = useAppSelector((state) => state.ocr.imageProvider)
+  const [provider, setProvider] = useState<OcrProvider>(imageProvider) // since default to image provider

  const tabs: TabsProps['items'] = [
    {
      key: 'image',
      label: t('settings.tool.ocr.image.title'),
      icon: <PictureOutlined />,
-      children: <OcrImageProviderSettings />
+      children: <OcrImageSettings setProvider={setProvider} />
    }
  ]

@ -27,6 +32,9 @@ const OcrSettings: FC = () => {
        <SettingDivider />
        <Tabs defaultActiveKey="image" items={tabs} />
      </SettingGroup>
+      <SettingGroup theme={themeMode}>
+        <OcrProviderSettings provider={provider} />
+      </SettingGroup>
    </>
  )
 }
--- a/src/renderer/src/pages/settings/DocProcessSettings/OcrTesseractSettings.tsx
+++ b/src/renderer/src/pages/settings/DocProcessSettings/OcrTesseractSettings.tsx
@ -0,0 +1,52 @@
+// import { loggerService } from '@logger'
+import InfoTooltip from '@renderer/components/InfoTooltip'
+import { useOcrProvider } from '@renderer/hooks/useOcrProvider'
+import { BuiltinOcrProviderIds, isOcrTesseractProvider } from '@renderer/types'
+import { Flex, Select } from 'antd'
+import { useTranslation } from 'react-i18next'
+
+import { SettingRow, SettingRowTitle } from '..'
+
+// const logger = loggerService.withContext('OcrTesseractSettings')
+
+export const OcrTesseractSettings = () => {
+  const { t } = useTranslation()
+  const { provider } = useOcrProvider(BuiltinOcrProviderIds.tesseract)
+
+  // TODO: use error boundary
+  if (!isOcrTesseractProvider(provider)) {
+    throw new Error('Not tesseract provider.')
+  }
+
+  // const [langs, setLangs] = useState<OcrTesseractConfig['langs']>(provider.config?.langs ?? {})
+
+  // currently static
+  const options = [
+    { value: 'chi_sim', label: t('languages.chinese') },
+    { value: 'chi_tra', label: t('languages.chinese-traditional') },
+    { value: 'eng', label: t('languages.english') }
+  ]
+
+  return (
+    <>
+      <SettingRow>
+        <SettingRowTitle>
+          <Flex align="center" gap={4}>
+            {t('settings.tool.ocr.image.tesseract.langs')}
+            <InfoTooltip title={t('settings.tool.ocr.image.tesseract.temp_tooltip')} />
+          </Flex>
+        </SettingRowTitle>
+        <div style={{ display: 'flex', gap: '8px' }}>
+          <Select
+            mode="multiple"
+            disabled
+            style={{ width: '100%' }}
+            placeholder="Please select"
+            value={['chi_sim', 'chi_tra', 'eng']}
+            options={options}
+          />
+        </div>
+      </SettingRow>
+    </>
+  )
+}
--- a/src/renderer/src/store/ocr.ts
+++ b/src/renderer/src/store/ocr.ts
@ -1,6 +1,6 @@
 import { createSlice, PayloadAction } from '@reduxjs/toolkit'
 import { BUILTIN_OCR_PROVIDERS, DEFAULT_OCR_PROVIDER } from '@renderer/config/ocr'
-import { ImageOcrProvider, OcrProvider } from '@renderer/types'
+import { ImageOcrProvider, OcrProvider, OcrProviderConfig } from '@renderer/types'

 export interface OcrState {
  providers: OcrProvider[]
@ -22,8 +22,8 @@ const ocrSlice = createSlice({
    addOcrProvider(state, action: PayloadAction<OcrProvider>) {
      state.providers.push(action.payload)
    },
-    removeOcrProvider(state, action: PayloadAction<OcrProvider>) {
-      state.providers = state.providers.filter((provider) => provider.id !== action.payload.id)
+    removeOcrProvider(state, action: PayloadAction<string>) {
+      state.providers = state.providers.filter((provider) => provider.id !== action.payload)
    },
    updateOcrProvider(state, action: PayloadAction<Partial<OcrProvider>>) {
      const index = state.providers.findIndex((provider) => provider.id === action.payload.id)
@ -31,13 +31,31 @@ const ocrSlice = createSlice({
        Object.assign(state.providers[index], action.payload)
      }
    },
+    updateOcrProviderConfig(
+      state,
+      action: PayloadAction<{ id: string; update: Omit<Partial<OcrProviderConfig>, 'id'> }>
+    ) {
+      const index = state.providers.findIndex((provider) => provider.id === action.payload.id)
+      if (index !== -1) {
+        if (!state.providers[index].config) {
+          state.providers[index].config = {}
+        }
+        Object.assign(state.providers[index].config, action.payload.update)
+      }
+    },
    setImageOcrProvider(state, action: PayloadAction<ImageOcrProvider>) {
      state.imageProvider = action.payload
    }
  }
 })

-export const { setOcrProviders, addOcrProvider, removeOcrProvider, updateOcrProvider, setImageOcrProvider } =
-  ocrSlice.actions
+export const {
+  setOcrProviders,
+  addOcrProvider,
+  removeOcrProvider,
+  updateOcrProvider,
+  updateOcrProviderConfig,
+  setImageOcrProvider
+} = ocrSlice.actions

 export default ocrSlice.reducer
--- a/src/renderer/src/types/ocr.ts
+++ b/src/renderer/src/types/ocr.ts
@ -1,3 +1,5 @@
+import Tesseract from 'tesseract.js'
+
 import { FileMetadata, ImageFileMetadata, isImageFile } from '.'

 export const BuiltinOcrProviderIds = {
@ -123,3 +125,18 @@ export type OcrResult = {
 export type OcrHandler = (file: SupportedOcrFile) => Promise<OcrResult>

 export type OcrImageHandler = (file: ImageFileMetadata) => Promise<OcrResult>
+
+// Tesseract Types
+export type OcrTesseractConfig = OcrProviderConfig & {
+  langs: Record<TesseractLangCode, boolean>
+}
+
+export type OcrTesseractProvider = BuiltinOcrProvider & {
+  config: OcrTesseractConfig
+}
+
+export const isOcrTesseractProvider = (p: OcrProvider): p is OcrTesseractProvider => {
+  return p.id === BuiltinOcrProviderIds.tesseract
+}
+
+export type TesseractLangCode = Tesseract.LanguageCode
--- a/yarn.lock
+++ b/yarn.lock
@ -8579,7 +8579,7 @@ __metadata:
    string-width: "npm:^7.2.0"
    styled-components: "npm:^6.1.11"
    tar: "npm:^7.4.3"
-    tesseract.js: "npm:^6.0.1"
+    tesseract.js: "patch:tesseract.js@npm%3A6.0.1#~/.yarn/patches/tesseract.js-npm-6.0.1-2562a7e46d.patch"
    tiny-pinyin: "npm:^1.3.2"
    tokenx: "npm:^1.1.0"
    tsx: "npm:^4.20.3"
@ -20978,7 +20978,7 @@ __metadata:
  languageName: node
  linkType: hard

-"tesseract.js@npm:*, tesseract.js@npm:^6.0.1":
+"tesseract.js@npm:6.0.1":
  version: 6.0.1
  resolution: "tesseract.js@npm:6.0.1"
  dependencies:
@ -20995,6 +20995,23 @@ __metadata:
  languageName: node
  linkType: hard

+"tesseract.js@patch:tesseract.js@npm%3A6.0.1#~/.yarn/patches/tesseract.js-npm-6.0.1-2562a7e46d.patch":
+  version: 6.0.1
+  resolution: "tesseract.js@patch:tesseract.js@npm%3A6.0.1#~/.yarn/patches/tesseract.js-npm-6.0.1-2562a7e46d.patch::version=6.0.1&hash=a9cf7b"
+  dependencies:
+    bmp-js: "npm:^0.1.0"
+    idb-keyval: "npm:^6.2.0"
+    is-url: "npm:^1.2.4"
+    node-fetch: "npm:^2.6.9"
+    opencollective-postinstall: "npm:^2.0.3"
+    regenerator-runtime: "npm:^0.13.3"
+    tesseract.js-core: "npm:^6.0.0"
+    wasm-feature-detect: "npm:^1.2.11"
+    zlibjs: "npm:^0.3.1"
+  checksum: 10c0/8a94fcc688ff21a9e82b721563d8fa174837ba807d0f01290fe9a1bb6a1c96ecaf7dc1c83510510f3d5185abd15f1cc5fc3cb7ad6c0eee0c4b3e278106f8a5da
+  languageName: node
+  linkType: hard
+
 "test-exclude@npm:^7.0.1":
  version: 7.0.1
  resolution: "test-exclude@npm:7.0.1"