LaTeX2Unicode translates LaTeX markup to human readable Unicode when possible. Here's an online demo that can be conveniently used to type in special characters. (demo source)

Basic math notations are supported. For instance:

\because \t{AB} + \t{BC} \neq \t{AC}
\therefore \iint\sqrt[4]{\xi^{\theta+1}} - \frac 38 \le
\Sigma \zeta_i \\
\therefore \exists{x}\forall{y} x \in \^A

is converted to

∵ A͡B + B͡C ≠ A͡C ∴ ∬∜ξ̅ᶿ̅⁺̅¹̅ - ⅜ ≤ Σ ζᵢ

∴ ∃x∀y x ∈ Â

Hundreds of other symbols and special characters are supported, too. For example, \spadesuit, \aleph, \OE, \downdownarrows and \o are translated to ♠, ℵ, Œ, ⇊, ø, respectively.

Some font styles are supported, too. For instance:

\textbb{Black Board Bold}, \textfrak{Fraktur},
{\bf Bold Face}, {\cal Calligraphic}, {\it Italic},
{\tt Monospace}

is translated to

𝔹𝕝𝕒𝕔𝕜 𝔹𝕠𝕒𝕣𝕕 𝔹𝕠𝕝𝕕, 𝔉𝔯𝔞𝔨𝔱𝔲𝔯, 𝐁𝐨𝐥𝐝 𝐅𝐚𝐜𝐞, 𝓒𝓪𝓵𝓵𝓲𝓰𝓻𝓪𝓹𝓱𝓲𝓬, 𝐼𝑡𝑎𝑙𝑖𝑐, 𝙼𝚘𝚗𝚘𝚜𝚙𝚊𝚌𝚎

LaTeX2Unicode is written in Scala, thus can serve as a 3rd party library in any JVM application that needs to extract information from LaTeX texts (e.g. BibTeX).

For simple conversion without configuration, which works fine in most cases, one call to a static method and you're done.

In Scala:

import com.github.tomtung.latex2unicode._

val latex = "\\ss^2 + \\alpha_3 \n \\div \\frac{1}{3} = \\sqrt[3]{123}"
val unicode = LaTeX2Unicode.convert(latex)
println(unicode)

In Java:

import com.github.tomtung.latex2unicode.LaTeX2Unicode;

String latex = "\\ss^2 + \\alpha_3 \n \\div \\frac{1}{3} = \\sqrt[3]{123}"
String unicode = LaTeX2Unicode.convert(latex)
System.out.println(unicode);

To add dependency on LaTeX2Unicode, insert the following to your pom.xml file if you use Apache Maven:

<dependency>
    <groupId>com.github.tomtung</groupId>
    <artifactId>latex2unicode_2.12</artifactId>
    <version>0.3.2</version>
</dependency>

or add the following to your build.sbt file if you use sbt 0.11+:

libraryDependencies += "com.github.tomtung" %% "latex2unicode" % "0.3.2"

LaTeX2Unicode is inspired by two similar projects, latex-to-unicode by ypsu (written in Python) and latex-to-unicode by vikhyat (written in Ruby).

LaTeX2Unicode is built on fastparse, an fast and elegant PEG parsing framework.

Apache License Version 2.0