compromise-speech
pronounciation plugin for compromise
npm install compromise-speech
### Syllables
Tokenize a word approximately by it's pronounced syllables.
We use a custom rule-based alortithm for this, to split the words into intuitive, readable syllable chunks.
```js
import plg from 'compromise-speech'
nlp.extend(plg)
const doc = nlp(`Albertville, Saskatchewan`)
doc.syllables()
/*
[
[
'Al',
'bert',
'ville,',
'Sa',
'skat',
'che',
'wan'
]
]
*/
```
Results are returned as an array of sentences, split by syllable.
If you want to return the individual words, or any other custom splitting:
```js
const doc = nlp(`Calgary Flames`)
const byTerm = doc.terms().syllables()
// [ [ 'Cal', 'ga', 'ry' ], [ 'Flames' ] ]
const doc = nlp(`it stars Alicia Silverstone and Paul Rudd`)
const pronounceNames = doc.people().syllables()
//[
// [ 'A', 'li', 'cia', 'Sil', 'ver', 'stone' ],
// [ 'Paul', 'Rudd' ]
// ]
```
you can have it operate on normalized words as well:
```js
const doc = nlp(`BACK to Winnipeg!`)
doc.syllables({ normalize: true })
// [ [ 'back', 'to', 'win', 'ni', 'peg' ] ]
```
if you'd like to combine syllable information with other properties, you can use `.compute('syllables')`
```js
let doc = nlp('edmonton oilers')
doc.compute('syllables')
const json = doc.json()
/*[{
"terms": [
{
"normal": "edmonton",
"syllables": ["ed", "mon", "ton"]
},
{
"normal": "oilers",
"syllables": ["oi", "lers"]
}
]
}]
*/
```
### Note
Pronunciation is regional, cultural, and even varies person-to-person. Some efforts have been made to support a wide variety of linguistic interpretations, but this may not be very wide indeed.
Please refer to more sophisticated libraries if your application requires it, or help us improve.
### See also
- [Talisman syllable tokenizers](https://yomguithereal.github.io/talisman/tokenizers/syllables) (JavaScript) — returns word chunks using SonoriPy's sonority rules or LegaliPy's onset rules learned from a text corpus. A close comparison for splitting the original spelling.
- [NLTK SyllableTokenizer](https://www.nltk.org/api/nltk.tokenize.sonority_sequencing.html) (Python) — splits words using the Sonority Sequencing Principle, with an English alphabet hierarchy by default; for example, `justification` becomes `jus / ti / fi / ca / tion`.
- [words/syllable](https://github.com/words/syllable) (JavaScript) — estimates English syllable counts, returning a number rather than word chunks.
- [hyphen](https://github.com/ytiurin/hyphen) (JavaScript) — uses Liang's hyphenation algorithm and language-specific patterns to insert line-break opportunities into written words.
- [Hyphenopoly](https://github.com/mnater/Hyphenopoly) (JavaScript/WebAssembly) — hyphenation for browsers and Node. Its [syllabification notes](https://mnater.github.io/Hyphenopoly/Special-use-cases.html#syllabification) explain why line breaks differ from syllable boundaries and describe a German-specific syllable pattern set.
MIT