react-native-executorch 0.5.1 → 0.5.2

package/lib/modules/natural_language_processing/TextEmbeddingsModule.js

@@ -0,0 +1,21 @@
+import { ResourceFetcher } from '../../utils/ResourceFetcher';
+import { BaseNonStaticModule } from '../BaseNonStaticModule';
+export class TextEmbeddingsModule extends BaseNonStaticModule {
+    async load(model, onDownloadProgressCallback = () => { }) {
+        const modelPromise = ResourceFetcher.fetch(onDownloadProgressCallback, model.modelSource);
+        const tokenizerPromise = ResourceFetcher.fetch(undefined, model.tokenizerSource);
+        const [modelResult, tokenizerResult] = await Promise.all([
+            modelPromise,
+            tokenizerPromise,
+        ]);
+        const modelPath = modelResult?.[0];
+        const tokenizerPath = tokenizerResult?.[0];
+        if (!modelPath || !tokenizerPath) {
+            throw new Error('Download interrupted.');
+        }
+        this.nativeModule = global.loadTextEmbeddings(modelPath, tokenizerPath);
+    }
+    async forward(input) {
+        return new Float32Array(await this.nativeModule.generate(input));
+    }
+}

package/lib/modules/natural_language_processing/TokenizerModule.d.ts

@@ -0,0 +1,12 @@
+import { ResourceSource } from '../../types/common';
+export declare class TokenizerModule {
+    nativeModule: any;
+    load(tokenizer: {
+        tokenizerSource: ResourceSource;
+    }, onDownloadProgressCallback?: (progress: number) => void): Promise<void>;
+    encode(s: string): Promise<any>;
+    decode(tokens: number[], skipSpecialTokens?: boolean): Promise<any>;
+    getVocabSize(): Promise<number>;
+    idToToken(tokenId: number): Promise<string>;
+    tokenToId(token: string): Promise<number>;
+}

package/lib/modules/natural_language_processing/TokenizerModule.js

@@ -0,0 +1,30 @@
+import { ResourceFetcher } from '../../utils/ResourceFetcher';
+export class TokenizerModule {
+    nativeModule;
+    async load(tokenizer, onDownloadProgressCallback = () => { }) {
+        const paths = await ResourceFetcher.fetch(onDownloadProgressCallback, tokenizer.tokenizerSource);
+        const path = paths?.[0];
+        if (!path) {
+            throw new Error('Download interrupted.');
+        }
+        this.nativeModule = global.loadTokenizerModule(path);
+    }
+    async encode(s) {
+        return await this.nativeModule.encode(s);
+    }
+    async decode(tokens, skipSpecialTokens = true) {
+        if (tokens.length === 0) {
+            return '';
+        }
+        return await this.nativeModule.decode(tokens, skipSpecialTokens);
+    }
+    async getVocabSize() {
+        return await this.nativeModule.getVocabSize();
+    }
+    async idToToken(tokenId) {
+        return this.nativeModule.idToToken(tokenId);
+    }
+    async tokenToId(token) {
+        return await this.nativeModule.tokenToId(token);
+    }
+}

package/lib/native/NativeETInstaller.d.ts

@@ -0,0 +1,6 @@
+import type { TurboModule } from 'react-native';
+export interface Spec extends TurboModule {
+    install(): boolean;
+}
+declare const _default: Spec | null;
+export default _default;

package/lib/native/NativeETInstaller.js

@@ -0,0 +1,2 @@
+import { TurboModuleRegistry } from 'react-native';
+export default TurboModuleRegistry.get('ETInstaller');

package/lib/native/NativeOCR.d.ts

@@ -0,0 +1,8 @@
+import type { TurboModule } from 'react-native';
+import { OCRDetection } from '../types/ocr';
+export interface Spec extends TurboModule {
+    loadModule(detectorSource: string, recognizerSourceLarge: string, recognizerSourceMedium: string, recognizerSourceSmall: string, symbols: string): Promise<number>;
+    forward(input: string): Promise<OCRDetection[]>;
+}
+declare const _default: Spec | null;
+export default _default;

package/lib/native/NativeOCR.js

@@ -0,0 +1,2 @@
+import { TurboModuleRegistry } from 'react-native';
+export default TurboModuleRegistry.get('OCR');

package/lib/native/NativeVerticalOCR.d.ts

@@ -0,0 +1,8 @@
+import type { TurboModule } from 'react-native';
+import { OCRDetection } from '../types/ocr';
+export interface Spec extends TurboModule {
+    loadModule(detectorLargeSource: string, detectorNarrowSource: string, recognizerSource: string, symbols: string, independentCharacters: boolean): Promise<number>;
+    forward(input: string): Promise<OCRDetection[]>;
+}
+declare const _default: Spec | null;
+export default _default;

package/lib/native/NativeVerticalOCR.js

@@ -0,0 +1,2 @@
+import { TurboModuleRegistry } from 'react-native';
+export default TurboModuleRegistry.get('VerticalOCR');

package/lib/native/RnExecutorchModules.d.ts

@@ -0,0 +1,3 @@
+import { Spec as ETInstallerInterface } from './NativeETInstaller';
+declare const ETInstallerNativeModule: ETInstallerInterface;
+export { ETInstallerNativeModule };

package/lib/native/RnExecutorchModules.js

@@ -0,0 +1,16 @@
+import { Platform } from 'react-native';
+const LINKING_ERROR = `The package 'react-native-executorch' doesn't seem to be linked. Make sure: \n\n` +
+    Platform.select({ ios: "- You have run 'pod install'\n", default: '' }) +
+    '- You rebuilt the app after installing the package\n' +
+    '- You are not using Expo Go\n';
+function returnSpecOrThrowLinkingError(spec) {
+    return spec
+        ? spec
+        : new Proxy({}, {
+            get() {
+                throw new Error(LINKING_ERROR);
+            },
+        });
+}
+const ETInstallerNativeModule = returnSpecOrThrowLinkingError(require('./NativeETInstaller').default);
+export { ETInstallerNativeModule };

package/lib/types/common.d.ts

@@ -0,0 +1,31 @@
+export type ResourceSource = string | number | object;
+export declare enum ScalarType {
+    BYTE = 0,
+    CHAR = 1,
+    SHORT = 2,
+    INT = 3,
+    LONG = 4,
+    HALF = 5,
+    FLOAT = 6,
+    DOUBLE = 7,
+    BOOL = 11,
+    QINT8 = 12,
+    QUINT8 = 13,
+    QINT32 = 14,
+    QUINT4X2 = 16,
+    QUINT2X4 = 17,
+    BITS16 = 22,
+    FLOAT8E5M2 = 23,
+    FLOAT8E4M3FN = 24,
+    FLOAT8E5M2FNUZ = 25,
+    FLOAT8E4M3FNUZ = 26,
+    UINT16 = 27,
+    UINT32 = 28,
+    UINT64 = 29
+}
+export type TensorBuffer = ArrayBuffer | Float32Array | Float64Array | Int8Array | Int16Array | Int32Array | Uint8Array | Uint16Array | Uint32Array | BigInt64Array | BigUint64Array;
+export interface TensorPtr {
+    dataPtr: TensorBuffer;
+    sizes: number[];
+    scalarType: ScalarType;
+}

package/lib/types/common.js

@@ -0,0 +1,25 @@
+export var ScalarType;
+(function (ScalarType) {
+    ScalarType[ScalarType["BYTE"] = 0] = "BYTE";
+    ScalarType[ScalarType["CHAR"] = 1] = "CHAR";
+    ScalarType[ScalarType["SHORT"] = 2] = "SHORT";
+    ScalarType[ScalarType["INT"] = 3] = "INT";
+    ScalarType[ScalarType["LONG"] = 4] = "LONG";
+    ScalarType[ScalarType["HALF"] = 5] = "HALF";
+    ScalarType[ScalarType["FLOAT"] = 6] = "FLOAT";
+    ScalarType[ScalarType["DOUBLE"] = 7] = "DOUBLE";
+    ScalarType[ScalarType["BOOL"] = 11] = "BOOL";
+    ScalarType[ScalarType["QINT8"] = 12] = "QINT8";
+    ScalarType[ScalarType["QUINT8"] = 13] = "QUINT8";
+    ScalarType[ScalarType["QINT32"] = 14] = "QINT32";
+    ScalarType[ScalarType["QUINT4X2"] = 16] = "QUINT4X2";
+    ScalarType[ScalarType["QUINT2X4"] = 17] = "QUINT2X4";
+    ScalarType[ScalarType["BITS16"] = 22] = "BITS16";
+    ScalarType[ScalarType["FLOAT8E5M2"] = 23] = "FLOAT8E5M2";
+    ScalarType[ScalarType["FLOAT8E4M3FN"] = 24] = "FLOAT8E4M3FN";
+    ScalarType[ScalarType["FLOAT8E5M2FNUZ"] = 25] = "FLOAT8E5M2FNUZ";
+    ScalarType[ScalarType["FLOAT8E4M3FNUZ"] = 26] = "FLOAT8E4M3FNUZ";
+    ScalarType[ScalarType["UINT16"] = 27] = "UINT16";
+    ScalarType[ScalarType["UINT32"] = 28] = "UINT32";
+    ScalarType[ScalarType["UINT64"] = 29] = "UINT64";
+})(ScalarType || (ScalarType = {}));

package/lib/types/imageSegmentation.d.ts

@@ -0,0 +1,24 @@
+export declare enum DeeplabLabel {
+    BACKGROUND = 0,
+    AEROPLANE = 1,
+    BICYCLE = 2,
+    BIRD = 3,
+    BOAT = 4,
+    BOTTLE = 5,
+    BUS = 6,
+    CAR = 7,
+    CAT = 8,
+    CHAIR = 9,
+    COW = 10,
+    DININGTABLE = 11,
+    DOG = 12,
+    HORSE = 13,
+    MOTORBIKE = 14,
+    PERSON = 15,
+    POTTEDPLANT = 16,
+    SHEEP = 17,
+    SOFA = 18,
+    TRAIN = 19,
+    TVMONITOR = 20,
+    ARGMAX = 21
+}

package/lib/types/imageSegmentation.js

@@ -0,0 +1,26 @@
+/* eslint-disable @cspell/spellchecker */
+export var DeeplabLabel;
+(function (DeeplabLabel) {
+    DeeplabLabel[DeeplabLabel["BACKGROUND"] = 0] = "BACKGROUND";
+    DeeplabLabel[DeeplabLabel["AEROPLANE"] = 1] = "AEROPLANE";
+    DeeplabLabel[DeeplabLabel["BICYCLE"] = 2] = "BICYCLE";
+    DeeplabLabel[DeeplabLabel["BIRD"] = 3] = "BIRD";
+    DeeplabLabel[DeeplabLabel["BOAT"] = 4] = "BOAT";
+    DeeplabLabel[DeeplabLabel["BOTTLE"] = 5] = "BOTTLE";
+    DeeplabLabel[DeeplabLabel["BUS"] = 6] = "BUS";
+    DeeplabLabel[DeeplabLabel["CAR"] = 7] = "CAR";
+    DeeplabLabel[DeeplabLabel["CAT"] = 8] = "CAT";
+    DeeplabLabel[DeeplabLabel["CHAIR"] = 9] = "CHAIR";
+    DeeplabLabel[DeeplabLabel["COW"] = 10] = "COW";
+    DeeplabLabel[DeeplabLabel["DININGTABLE"] = 11] = "DININGTABLE";
+    DeeplabLabel[DeeplabLabel["DOG"] = 12] = "DOG";
+    DeeplabLabel[DeeplabLabel["HORSE"] = 13] = "HORSE";
+    DeeplabLabel[DeeplabLabel["MOTORBIKE"] = 14] = "MOTORBIKE";
+    DeeplabLabel[DeeplabLabel["PERSON"] = 15] = "PERSON";
+    DeeplabLabel[DeeplabLabel["POTTEDPLANT"] = 16] = "POTTEDPLANT";
+    DeeplabLabel[DeeplabLabel["SHEEP"] = 17] = "SHEEP";
+    DeeplabLabel[DeeplabLabel["SOFA"] = 18] = "SOFA";
+    DeeplabLabel[DeeplabLabel["TRAIN"] = 19] = "TRAIN";
+    DeeplabLabel[DeeplabLabel["TVMONITOR"] = 20] = "TVMONITOR";
+    DeeplabLabel[DeeplabLabel["ARGMAX"] = 21] = "ARGMAX";
+})(DeeplabLabel || (DeeplabLabel = {}));

package/lib/types/llm.d.ts

@@ -0,0 +1,46 @@
+export interface LLMType {
+    messageHistory: Message[];
+    response: string;
+    token: string;
+    isReady: boolean;
+    isGenerating: boolean;
+    downloadProgress: number;
+    error: string | null;
+    configure: ({ chatConfig, toolsConfig, }: {
+        chatConfig?: Partial<ChatConfig>;
+        toolsConfig?: ToolsConfig;
+    }) => void;
+    generate: (messages: Message[], tools?: LLMTool[]) => Promise<void>;
+    sendMessage: (message: string) => Promise<void>;
+    deleteMessage: (index: number) => void;
+    interrupt: () => void;
+}
+export type MessageRole = 'user' | 'assistant' | 'system';
+export interface Message {
+    role: MessageRole;
+    content: string;
+}
+export interface ToolCall {
+    toolName: string;
+    arguments: Object;
+}
+export type LLMTool = Object;
+export interface ChatConfig {
+    initialMessageHistory: Message[];
+    contextWindowLength: number;
+    systemPrompt: string;
+}
+export interface ToolsConfig {
+    tools: LLMTool[];
+    executeToolCallback: (call: ToolCall) => Promise<string | null>;
+    displayToolCalls?: boolean;
+}
+export declare const SPECIAL_TOKENS: {
+    BOS_TOKEN: string;
+    EOS_TOKEN: string;
+    UNK_TOKEN: string;
+    SEP_TOKEN: string;
+    PAD_TOKEN: string;
+    CLS_TOKEN: string;
+    MASK_TOKEN: string;
+};

package/lib/types/llm.js

@@ -0,0 +1,9 @@
+export const SPECIAL_TOKENS = {
+    BOS_TOKEN: 'bos_token',
+    EOS_TOKEN: 'eos_token',
+    UNK_TOKEN: 'unk_token',
+    SEP_TOKEN: 'sep_token',
+    PAD_TOKEN: 'pad_token',
+    CLS_TOKEN: 'cls_token',
+    MASK_TOKEN: 'mask_token',
+};

package/lib/types/objectDetection.d.ts

@@ -0,0 +1,104 @@
+export interface Bbox {
+    x1: number;
+    x2: number;
+    y1: number;
+    y2: number;
+}
+export interface Detection {
+    bbox: Bbox;
+    label: keyof typeof CocoLabel;
+    score: number;
+}
+declare enum CocoLabel {
+    PERSON = 1,
+    BICYCLE = 2,
+    CAR = 3,
+    MOTORCYCLE = 4,
+    AIRPLANE = 5,
+    BUS = 6,
+    TRAIN = 7,
+    TRUCK = 8,
+    BOAT = 9,
+    TRAFFIC_LIGHT = 10,
+    FIRE_HYDRANT = 11,
+    STREET_SIGN = 12,
+    STOP_SIGN = 13,
+    PARKING = 14,
+    BENCH = 15,
+    BIRD = 16,
+    CAT = 17,
+    DOG = 18,
+    HORSE = 19,
+    SHEEP = 20,
+    COW = 21,
+    ELEPHANT = 22,
+    BEAR = 23,
+    ZEBRA = 24,
+    GIRAFFE = 25,
+    HAT = 26,
+    BACKPACK = 27,
+    UMBRELLA = 28,
+    SHOE = 29,
+    EYE = 30,
+    HANDBAG = 31,
+    TIE = 32,
+    SUITCASE = 33,
+    FRISBEE = 34,
+    SKIS = 35,
+    SNOWBOARD = 36,
+    SPORTS = 37,
+    KITE = 38,
+    BASEBALL = 39,
+    SKATEBOARD = 41,
+    SURFBOARD = 42,
+    TENNIS_RACKET = 43,
+    BOTTLE = 44,
+    PLATE = 45,
+    WINE_GLASS = 46,
+    CUP = 47,
+    FORK = 48,
+    KNIFE = 49,
+    SPOON = 50,
+    BOWL = 51,
+    BANANA = 52,
+    APPLE = 53,
+    SANDWICH = 54,
+    ORANGE = 55,
+    BROCCOLI = 56,
+    CARROT = 57,
+    HOT_DOG = 58,
+    PIZZA = 59,
+    DONUT = 60,
+    CAKE = 61,
+    CHAIR = 62,
+    COUCH = 63,
+    POTTED_PLANT = 64,
+    BED = 65,
+    MIRROR = 66,
+    DINING_TABLE = 67,
+    WINDOW = 68,
+    DESK = 69,
+    TOILET = 70,
+    DOOR = 71,
+    TV = 72,
+    LAPTOP = 73,
+    MOUSE = 74,
+    REMOTE = 75,
+    KEYBOARD = 76,
+    CELL_PHONE = 77,
+    MICROWAVE = 78,
+    OVEN = 79,
+    TOASTER = 80,
+    SINK = 81,
+    REFRIGERATOR = 82,
+    BLENDER = 83,
+    BOOK = 84,
+    CLOCK = 85,
+    VASE = 86,
+    SCISSORS = 87,
+    TEDDY_BEAR = 88,
+    HAIR_DRIER = 89,
+    TOOTHBRUSH = 90,
+    HAIR_BRUSH = 91
+}
+export {};

package/lib/types/objectDetection.js

@@ -0,0 +1,94 @@
+var CocoLabel;
+(function (CocoLabel) {
+    CocoLabel[CocoLabel["PERSON"] = 1] = "PERSON";
+    CocoLabel[CocoLabel["BICYCLE"] = 2] = "BICYCLE";
+    CocoLabel[CocoLabel["CAR"] = 3] = "CAR";
+    CocoLabel[CocoLabel["MOTORCYCLE"] = 4] = "MOTORCYCLE";
+    CocoLabel[CocoLabel["AIRPLANE"] = 5] = "AIRPLANE";
+    CocoLabel[CocoLabel["BUS"] = 6] = "BUS";
+    CocoLabel[CocoLabel["TRAIN"] = 7] = "TRAIN";
+    CocoLabel[CocoLabel["TRUCK"] = 8] = "TRUCK";
+    CocoLabel[CocoLabel["BOAT"] = 9] = "BOAT";
+    CocoLabel[CocoLabel["TRAFFIC_LIGHT"] = 10] = "TRAFFIC_LIGHT";
+    CocoLabel[CocoLabel["FIRE_HYDRANT"] = 11] = "FIRE_HYDRANT";
+    CocoLabel[CocoLabel["STREET_SIGN"] = 12] = "STREET_SIGN";
+    CocoLabel[CocoLabel["STOP_SIGN"] = 13] = "STOP_SIGN";
+    CocoLabel[CocoLabel["PARKING"] = 14] = "PARKING";
+    CocoLabel[CocoLabel["BENCH"] = 15] = "BENCH";
+    CocoLabel[CocoLabel["BIRD"] = 16] = "BIRD";
+    CocoLabel[CocoLabel["CAT"] = 17] = "CAT";
+    CocoLabel[CocoLabel["DOG"] = 18] = "DOG";
+    CocoLabel[CocoLabel["HORSE"] = 19] = "HORSE";
+    CocoLabel[CocoLabel["SHEEP"] = 20] = "SHEEP";
+    CocoLabel[CocoLabel["COW"] = 21] = "COW";
+    CocoLabel[CocoLabel["ELEPHANT"] = 22] = "ELEPHANT";
+    CocoLabel[CocoLabel["BEAR"] = 23] = "BEAR";
+    CocoLabel[CocoLabel["ZEBRA"] = 24] = "ZEBRA";
+    CocoLabel[CocoLabel["GIRAFFE"] = 25] = "GIRAFFE";
+    CocoLabel[CocoLabel["HAT"] = 26] = "HAT";
+    CocoLabel[CocoLabel["BACKPACK"] = 27] = "BACKPACK";
+    CocoLabel[CocoLabel["UMBRELLA"] = 28] = "UMBRELLA";
+    CocoLabel[CocoLabel["SHOE"] = 29] = "SHOE";
+    CocoLabel[CocoLabel["EYE"] = 30] = "EYE";
+    CocoLabel[CocoLabel["HANDBAG"] = 31] = "HANDBAG";
+    CocoLabel[CocoLabel["TIE"] = 32] = "TIE";
+    CocoLabel[CocoLabel["SUITCASE"] = 33] = "SUITCASE";
+    CocoLabel[CocoLabel["FRISBEE"] = 34] = "FRISBEE";
+    CocoLabel[CocoLabel["SKIS"] = 35] = "SKIS";
+    CocoLabel[CocoLabel["SNOWBOARD"] = 36] = "SNOWBOARD";
+    CocoLabel[CocoLabel["SPORTS"] = 37] = "SPORTS";
+    CocoLabel[CocoLabel["KITE"] = 38] = "KITE";
+    CocoLabel[CocoLabel["BASEBALL"] = 39] = "BASEBALL";
+    CocoLabel[CocoLabel["SKATEBOARD"] = 41] = "SKATEBOARD";
+    CocoLabel[CocoLabel["SURFBOARD"] = 42] = "SURFBOARD";
+    CocoLabel[CocoLabel["TENNIS_RACKET"] = 43] = "TENNIS_RACKET";
+    CocoLabel[CocoLabel["BOTTLE"] = 44] = "BOTTLE";
+    CocoLabel[CocoLabel["PLATE"] = 45] = "PLATE";
+    CocoLabel[CocoLabel["WINE_GLASS"] = 46] = "WINE_GLASS";
+    CocoLabel[CocoLabel["CUP"] = 47] = "CUP";
+    CocoLabel[CocoLabel["FORK"] = 48] = "FORK";
+    CocoLabel[CocoLabel["KNIFE"] = 49] = "KNIFE";
+    CocoLabel[CocoLabel["SPOON"] = 50] = "SPOON";
+    CocoLabel[CocoLabel["BOWL"] = 51] = "BOWL";
+    CocoLabel[CocoLabel["BANANA"] = 52] = "BANANA";
+    CocoLabel[CocoLabel["APPLE"] = 53] = "APPLE";
+    CocoLabel[CocoLabel["SANDWICH"] = 54] = "SANDWICH";
+    CocoLabel[CocoLabel["ORANGE"] = 55] = "ORANGE";
+    CocoLabel[CocoLabel["BROCCOLI"] = 56] = "BROCCOLI";
+    CocoLabel[CocoLabel["CARROT"] = 57] = "CARROT";
+    CocoLabel[CocoLabel["HOT_DOG"] = 58] = "HOT_DOG";
+    CocoLabel[CocoLabel["PIZZA"] = 59] = "PIZZA";
+    CocoLabel[CocoLabel["DONUT"] = 60] = "DONUT";
+    CocoLabel[CocoLabel["CAKE"] = 61] = "CAKE";
+    CocoLabel[CocoLabel["CHAIR"] = 62] = "CHAIR";
+    CocoLabel[CocoLabel["COUCH"] = 63] = "COUCH";
+    CocoLabel[CocoLabel["POTTED_PLANT"] = 64] = "POTTED_PLANT";
+    CocoLabel[CocoLabel["BED"] = 65] = "BED";
+    CocoLabel[CocoLabel["MIRROR"] = 66] = "MIRROR";
+    CocoLabel[CocoLabel["DINING_TABLE"] = 67] = "DINING_TABLE";
+    CocoLabel[CocoLabel["WINDOW"] = 68] = "WINDOW";
+    CocoLabel[CocoLabel["DESK"] = 69] = "DESK";
+    CocoLabel[CocoLabel["TOILET"] = 70] = "TOILET";
+    CocoLabel[CocoLabel["DOOR"] = 71] = "DOOR";
+    CocoLabel[CocoLabel["TV"] = 72] = "TV";
+    CocoLabel[CocoLabel["LAPTOP"] = 73] = "LAPTOP";
+    CocoLabel[CocoLabel["MOUSE"] = 74] = "MOUSE";
+    CocoLabel[CocoLabel["REMOTE"] = 75] = "REMOTE";
+    CocoLabel[CocoLabel["KEYBOARD"] = 76] = "KEYBOARD";
+    CocoLabel[CocoLabel["CELL_PHONE"] = 77] = "CELL_PHONE";
+    CocoLabel[CocoLabel["MICROWAVE"] = 78] = "MICROWAVE";
+    CocoLabel[CocoLabel["OVEN"] = 79] = "OVEN";
+    CocoLabel[CocoLabel["TOASTER"] = 80] = "TOASTER";
+    CocoLabel[CocoLabel["SINK"] = 81] = "SINK";
+    CocoLabel[CocoLabel["REFRIGERATOR"] = 82] = "REFRIGERATOR";
+    CocoLabel[CocoLabel["BLENDER"] = 83] = "BLENDER";
+    CocoLabel[CocoLabel["BOOK"] = 84] = "BOOK";
+    CocoLabel[CocoLabel["CLOCK"] = 85] = "CLOCK";
+    CocoLabel[CocoLabel["VASE"] = 86] = "VASE";
+    CocoLabel[CocoLabel["SCISSORS"] = 87] = "SCISSORS";
+    CocoLabel[CocoLabel["TEDDY_BEAR"] = 88] = "TEDDY_BEAR";
+    CocoLabel[CocoLabel["HAIR_DRIER"] = 89] = "HAIR_DRIER";
+    CocoLabel[CocoLabel["TOOTHBRUSH"] = 90] = "TOOTHBRUSH";
+    CocoLabel[CocoLabel["HAIR_BRUSH"] = 91] = "HAIR_BRUSH";
+})(CocoLabel || (CocoLabel = {}));
+export {};

package/lib/types/ocr.d.ts

@@ -0,0 +1,11 @@
+import { symbols } from '../constants/ocr/symbols';
+export interface OCRDetection {
+    bbox: OCRBbox[];
+    text: string;
+    score: number;
+}
+export interface OCRBbox {
+    x: number;
+    y: number;
+}
+export type OCRLanguage = keyof typeof symbols;

package/lib/types/ocr.js

@@ -0,0 +1 @@
+export {};

package/lib/types/stt.d.ts

@@ -0,0 +1,94 @@
+export interface ModelConfig {
+    sources: {
+        encoder: string;
+        decoder: string;
+    };
+    tokenizer: {
+        source: string;
+        bos: number;
+        eos: number;
+    };
+    isMultilingual: boolean;
+}
+export declare enum SpeechToTextLanguage {
+    Afrikaans = "af",
+    Albanian = "sq",
+    Arabic = "ar",
+    Armenian = "hy",
+    Azerbaijani = "az",
+    Basque = "eu",
+    Belarusian = "be",
+    Bengali = "bn",
+    Bosnian = "bs",
+    Bulgarian = "bg",
+    Burmese = "my",
+    Catalan = "ca",
+    Chinese = "zh",
+    Croatian = "hr",
+    Czech = "cs",
+    Danish = "da",
+    Dutch = "nl",
+    Estonian = "et",
+    English = "en",
+    Finnish = "fi",
+    French = "fr",
+    Galician = "gl",
+    Georgian = "ka",
+    German = "de",
+    Greek = "el",
+    Gujarati = "gu",
+    HaitianCreole = "ht",
+    Hebrew = "he",
+    Hindi = "hi",
+    Hungarian = "hu",
+    Icelandic = "is",
+    Indonesian = "id",
+    Italian = "it",
+    Japanese = "ja",
+    Kannada = "kn",
+    Kazakh = "kk",
+    Khmer = "km",
+    Korean = "ko",
+    Lao = "lo",
+    Latvian = "lv",
+    Lithuanian = "lt",
+    Macedonian = "mk",
+    Malagasy = "mg",
+    Malay = "ms",
+    Malayalam = "ml",
+    Maltese = "mt",
+    Marathi = "mr",
+    Nepali = "ne",
+    Norwegian = "no",
+    Persian = "fa",
+    Polish = "pl",
+    Portuguese = "pt",
+    Punjabi = "pa",
+    Romanian = "ro",
+    Russian = "ru",
+    Serbian = "sr",
+    Sinhala = "si",
+    Slovak = "sk",
+    Slovenian = "sl",
+    Spanish = "es",
+    Sundanese = "su",
+    Swahili = "sw",
+    Swedish = "sv",
+    Tagalog = "tl",
+    Tajik = "tg",
+    Tamil = "ta",
+    Telugu = "te",
+    Thai = "th",
+    Turkish = "tr",
+    Ukrainian = "uk",
+    Urdu = "ur",
+    Uzbek = "uz",
+    Vietnamese = "vi",
+    Welsh = "cy",
+    Yiddish = "yi"
+}
+export declare enum AvailableModels {
+    WHISPER = "whisper",
+    MOONSHINE = "moonshine",
+    WHISPER_MULTILINGUAL = "whisperMultilingual"
+}