phonemize/dist/index.js

Fast phonemizer with rule-based G2P prediction. Pure JavaScript implementation.

"use strict";
/**
 * Phonemize Library - Main API
 *
 * A comprehensive text-to-phoneme conversion library supporting:
 * - IPA (International Phonetic Alphabet) output
 * - ARPABET phonetic notation
 * - Multilingual text processing (Chinese, Japanese, Korean, Thai, Arabic, Russian)
 * - Rule-based G2P conversion with dictionary lookup
 * - Compound word decomposition
 * - Number and abbreviation expansion
 */
Object.defineProperty(exports, "__esModule", { value: true });
exports.Tokenizer = void 0;
exports.phonemize = phonemize;
exports.toIPA = toIPA;
exports.toARPABET = toARPABET;
exports.toZhuyin = toZhuyin;
exports.addPronunciation = addPronunciation;
exports.createTokenizer = createTokenizer;
const g2p_1 = require("./g2p");
const tokenizer_1 = require("./tokenizer");
Object.defineProperty(exports, "Tokenizer", { enumerable: true, get: function () { return tokenizer_1.Tokenizer; } });
/**
 * Main phonemize function implementation
 */
function phonemize(text, options = {}) {
    // Handle legacy boolean parameter
    if (options === true) {
        options = { returnArray: true };
    }
    const tokenizer = new tokenizer_1.Tokenizer(options);
    if (options.returnArray) {
        return tokenizer.tokenizeToTokens(text);
    }
    return tokenizer.tokenizeToString(text);
}
/**
 * Convert text to International Phonetic Alphabet (IPA) notation
 *
 * @param text - Input text to convert
 * @param options - Configuration options (format will be overridden to 'ipa')
 * @returns IPA phonetic string
 *
 * @example
 * ```typescript
 * toIPA("hello world") // "həloʊ wɝld"
 * toIPA("中文", { anyAscii: false }) // "ʈʂʊŋ˥˥ wən˧˥"
 * ```
 */
function toIPA(text, options) {
    const ipaOptions = Object.assign(Object.assign({}, options), { format: "ipa" });
    const tokenizer = new tokenizer_1.Tokenizer(ipaOptions);
    return tokenizer.tokenizeToString(text);
}
/**
 * Convert text to ARPABET phonetic notation
 *
 * @param text - Input text to convert
 * @param options - Configuration options (format will be overridden to 'arpabet')
 * @returns ARPABET phonetic string
 *
 * @example
 * ```typescript
 * toARPABET("hello world") // "HH AH L OW W ER L D"
 * toARPABET("testing", { stripStress: true }) // "T EH S T IH NG"
 * ```
 */
function toARPABET(text, options) {
    const arpabetOptions = Object.assign(Object.assign({}, options), { format: "arpabet" });
    const tokenizer = new tokenizer_1.Tokenizer(arpabetOptions);
    return tokenizer.tokenizeToString(text);
}
/**
 * Convert text to Zhuyin (Bopomofo) notation
 * Chinese characters are converted to Zhuyin with tone numbers,
 * non-Chinese characters are converted to IPA as fallback.
 *
 * @param text - Input text to convert
 * @param options - Configuration options (format will be overridden to 'zhuyin')
 * @returns Zhuyin phonetic string with tone numbers
 *
 * @example
 * ```typescript
 * toZhuyin("中文") // "ㄓㄨㄥ1 ㄨㄣ2"
 * toZhuyin("中文 hello") // "ㄓㄨㄥ1 ㄨㄣ2 həˈloʊ"
 * toZhuyin("測試", { stripStress: true }) // "ㄘㄜ4 ㄕ4"
 * ```
 */
function toZhuyin(text, options) {
    const zhuyinOptions = Object.assign(Object.assign({}, options), { format: "zhuyin" });
    const tokenizer = new tokenizer_1.Tokenizer(zhuyinOptions);
    return tokenizer.tokenizeToString(text);
}
/**
 * Add custom pronunciation to the internal dictionary
 *
 * @param word - Word to add pronunciation for
 * @param pronunciation - IPA pronunciation string
 *
 * @example
 * ```typescript
 * addPronunciation("github", "ɡɪthʌb");
 * toIPA("github") // "ɡɪthʌb"
 * ```
 */
function addPronunciation(word, pronunciation) {
    if (!(word === null || word === void 0 ? void 0 : word.trim()) || !(pronunciation === null || pronunciation === void 0 ? void 0 : pronunciation.trim())) {
        throw new Error("Both word and pronunciation must be non-empty strings");
    }
    g2p_1.g2pModel.addPronunciation(word.toLowerCase(), pronunciation);
}
/**
 * Create a custom tokenizer instance with specific configuration
 *
 * @param options - Tokenizer configuration options
 * @returns Configured Tokenizer instance
 *
 * @example
 * ```typescript
 * const tokenizer = createTokenizer({
 *   format: "ipa",
 *   stripStress: true,
 *   separator: "-"
 * });
 *
 * const result = tokenizer.tokenizeToString("hello");
 * ```
 */
function createTokenizer(options = {}) {
    return new tokenizer_1.Tokenizer(options);
}
/**
 * Phonemize library default export
 * Provides all core functions and classes for CommonJS compatibility
 */
const phonemizer = {
    // === Core Functions ===
    phonemize,
    toIPA,
    toARPABET,
    toZhuyin,
    // === Utilities ===
    addPronunciation,
    createTokenizer,
    // === Classes ===
    Tokenizer: tokenizer_1.Tokenizer,
};
exports.default = phonemizer;