LaTeX to Speech Conversion

Generates speech output for LaTeX expressions. This is based on MathJax and Speech Rule Engine.

Install

Install with

npm install latex-to-speech

Run on CLI

npx latex-to-speech [options] <expr...>

or

[PATH-TO-NODE_MODULES]/node_modules/.bin/latex-to-speech [options] <expr...>

Expressions are a list of LaTeX formulas. For options, see the options below. For more detailed explanations see also the documentation of Speech Rule Engine.

Use as Module

Loading the module returns the latexToSpeech function.

const latexToSpeech = require('latex-to-speech');

latexToSpeech takes a list of LaTeX expressions and a feature vector of options as input. It is an async function that returns a promise for the list of results. Here are some examples:

let result = await latexToSpeech(['x^2', '\\frac{a}{b}'], {});

will result in [ 'x squared', 'StartFraction a Over b EndFraction' ].

let result = await latexToSpeech(['x^2', '\\frac{a}{b}'], {domain: 'clearspeak'});

will result in [ 'x squared', 'a over b' ].

Options

Options for latex-to-speech are a subset of the Speech Rule Engine options. For more detailed explanations see also the documentation there.

| Option | Value | |------------|----------------------------------------------------------------------------------------------------------------| | domain | Domain or subject area of speech rules (e.g., mathspeak, clearspeak). | | style | Style or preference setting of speech rules (e.g., brief). | | | In case of clearspeak, multiple preferences can be chosen using : as separator. | | locale | Language locale in 639-1. | | markup | Set output markup for speech: none, ssml, sable, voicexml, acss, ssml_step | | modality | Set the modality SRE returns. E.g., speech, braille, prefix, summary |

Build

Building from github:

npm install; npm run build;