A set of Ithkuil tools, with more to come soon:
- IPA Translation
- Syllable Splitting (Beta)
npm install ithkuil-tools or yarn add ithkuil-tools.
Import the provided helper function:
import { romanizedIthkuilToIpa } from './lib/index'Then pass some romanized Ithkuil text to it to convert it to an IPA string representation:
const text = 'Wezvwaušburdóu yaizxra sai.' // "Be careful, your fork is actually a fennec."
const ipa: string | Error = romanizedIthkuilToIpa(text)Note the return type of romanizedIthkuilToIpa(): string | Error. This is because the conversion isn't guaranteed, either because of bad input or an internal bug. Just Add a check to be able to access the actual result:
if (typeof ipa === 'string') console.log(ipa) // [wɛzvwauʃbuɹd̪óu jaɪzxɾa sai]Conversely, you can inspect the returned error if that happens. It'll contain the error cause, what character index caused the failure, and the usual stack trace.
The full signature of the helper function is romanizedIthkuilToIpa(romanizedIthkuilText: string, options?: IpaConverterOptions): string | IpaConversionError. A few options can be provided, check the code for more details about those.
import {
romanizedIthkuilToSyllables,
romanizedIthkuilToSyllableBoundaries
} from "ithkuil-tools";
console.log(romanizedIthkuilToSyllables("Wezvwaušburdóu yaizxra sai")); // [ [ 'we', 'zvwauš', 'bur', 'dóu' ], [ 'yai', 'zxra' ], [ 'sai' ] ]
console.log(romanizedIthkuilToSyllableBoundaries("Wezvwaušburdóu yaizxra sai")) // [ [ 0, 2, 8, 11, 14 ], [ 0, 3, 7 ], [ 0, 3 ] ]IPA to speech is using AWS Polly behind the scenes, so you'll need to authenticate:
import { IpaToSpeech } from 'ithkuil-tools'
const ipaToSpeech = new IpaToSpeech({
credentials: {
accessKeyId: 'accessKeyId',
secretAccessKey: 'secretAccessKey',
},
region: 'region',
})Use the speak method, passing in some IPA text, to get a buffer containg the corresponding audio. Remember this is an async function returning a promise, so use it accordingly.
The following example should get you started, it saves the audio to an MP3 file:
import * as fs from 'fs'
ipaToSpeech.speak('hɛˈloʊ wɝld')
.then((buffer) => {
fs.writeFileSync('hello_world.mp3', buffer)
console.log(`Audio saved to file.`)
})