Type Greek letters in Math Mode using Greek keyboard [duplicate]

Question

I know that in order to type Greek letters in my equations I can use \alpha, \beta, etc. However, since I have a Greek keyboard, I would like to be able to write Greek characters right away α ,β, etc., which are just as easy as writing Latin ones. Is it possible? Do I have to redefine \alpha as α (how?), or is there a package to do that for me?

EDIT I'm also writing my text in Greek.

Answer 1

An answer for XeLaTeX (what I use sometimes). It can be also used in LaTeX but my MWE is for XeLaTeX:

% !TEX TS-program = XeLaTeX
% !TeX program = xelatex
\documentclass[a4paper,10pt]{article}

\usepackage{xltxtra}

\usepackage{polyglossia}
\setmainlanguage{greek}
\setotherlanguage{english}
\setmainfont[Mapping=tex-text]{Linux Libertine O}
% \setsansfont[Mapping=tex-text]{DejaVu Sans}
% \setmonofont[Mapping=tex-text]{DejaVu Sans Mono}

\newfontfamily\greekfont{Linux Libertine O}
\title{Τίτλος Άρθρου}
\author{Όνομα Επώνυμο}
\date{}

% ΓΙΑ ΕΛΛΗΝΙΚΑ ΣΤΑ ΜΑΘΗΜΑΤΙΚΑ
%Μικρά ελληνικά
\def\α{\alpha}
\def\β{\beta}
\def\γ{\gamma}
\def\δ{\delta}
\def\ε{\epsilon}
\def\ζ{\zeta}
\def\η{\eta}
\def\θ{\theta}
\def\ι{\iota}
\def\κ{\kappa}
\def\λ{\lambda}
\def\μ{\mu}
\def\ν{\nu}
\def\ξ{\xi}
\def\ο{o}
\def\π{\pi}
\def\ρ{\rho}
\def\σ{\sigma}
\def\τ{\tau}
\def\υ{\upsilon}
\def\φ{\varphi}
\def\χ{\chi}
\def\ψ{\psi}
\def\ω{\omega}
\def\ς{\varsigma}
\def\ά{\varalpha}

% Κεφαλαία Ελληνικά
\def\Α{\Alpha}
\def\Β{\Beta}
\def\Γ{\Gamma}
\def\Δ{\Delta}
\def\Ε{\Epsilon}
\def\Ζ{\Zeta}
\def\Η{\Eta}
\def\Θ{\Theta}
\def\Ι{\Iota}
\def\Κ{\Kappa}
\def\Λ{\Lambda}
\def\Μ{\Mu}
\def\Ν{\Nu}
\def\Ξ{\Xi}
\def\Ο{O}
\def\Π{\Pi}
\def\Ρ{\Rho}
\def\Σ{\Sigma}
\def\Τ{\Tau}
\def\Υ{\Ipsilon}
\def\Φ{\Phi}
\def\Χ{X}
\def\Ψ{\Psi}
\def\Ω{\Omega}
%ΤΕΛΟΣ ΓΙΑ ΕΛΛΗΝΙΚΑ ΣΤΑ ΜΑΘΗΜΑΤΙΚΑ

\begin{document}
\maketitle

\section{Ενότητα με μαθηματικά}

$f(x)=\α\cdot \ψ+\ο$

\begin{english}
   Text here more than a line that I hope it will be hyphenated if I add a BigEnglishWord
\end{english}

Ελληνικά που θα χρησιμοποιηθούν σαν βασικό κείμενο και θα έχουν τη δυνατότητα να κόβονται οι λέξεις

\end{document}

Output:

Edit: For another solution with XeLaTeX which uses our font in math to, take a look here: https://tex.stackexchange.com/a/396530/120578

Edit: And a LaTeX solution here without any definitions here:

\documentclass[a4paper,10pt]{article}
\usepackage[utf8]{inputenc}
\usepackage[english,greek]{babel}
\usepackage{alphabeta}

\usepackage[LGR, T1]{fontenc}


\title{Τίτλος Άρθρου}
\author{Όνομα Επώνυμο}
\date{}


\begin{document}
\maketitle

\section{Ενότητα με μαθηματικά}

$f(x)=a\cdot f+φ$

\begin{otherlanguage}{english}Text here more than a line that I hope it will be hyphenated if I add BigEnglishWord
\end{otherlanguage}

Ελληνικά που θα χρησιμοποιηθούν σαν βασικό κείμενο και θα έχουν τη δυνατότητα να κόβονται οι λέξεις

\end{document}

Source: https://www.sharelatex.com/learn/Greek (Please inform me if I need to add something because of sharelatex license... I can not find it there.)

Output:

Answer 2

Assuming that you are using pdflatex:

You can define the α so that it tests for math mode. But as the array in the example shows there are places where the test fails. Adding \relax before the \ifmmode would solve this, but destroy kerning in text mode. So a better solution would be a robust definition of α (David Carlisle could say more about this). You can get all the numbers from lgrenc.dfu.

\documentclass[]{article}

\usepackage[LGR]{fontenc}
\usepackage[utf8]{inputenc}


\DeclareUnicodeCharacter{03B1}{\ifmmode\alpha\else\textalpha\fi} % α
\DeclareUnicodeCharacter{03B2}{\ifmmode\beta\else\textbeta\fi}   % β


\begin{document}
α β

$α = β$

$
  \begin{array}{cc}
  α α & β β\\
  \end{array}
$
\end{document}

Edit

The rather new command \TextOrMath (it works from texlive 2015 on and is also used by the alphabeta package in koleygr's answer) actually avoids both problems so it (and the alphabeta package) is the better alternative to the primitive \ifmmode:

\documentclass[]{article}

\usepackage[LGR]{fontenc}
\usepackage[utf8]{inputenc}


\DeclareUnicodeCharacter{03B1}{\TextOrMath{\textalpha}{\alpha}} % α
\DeclareUnicodeCharacter{03B2}{\TextOrMath{\textbeta}{\beta}}   % β


\begin{document}
α β

$α = β$

$
  \begin{array}{cc}
  α α & β β\\
  \end{array}
$
\end{document}

Answer 3

If it doesn't clash with any other packages you load, you could use the following:

\documentclass[]{article}

\usepackage[mathletters]{ucs}
\usepackage[utf8x]{inputenc}

\begin{document}
$ω$ is prettier than $δ$
\end{document}

If the above clashes (and you use \usepackage[utf8]{inputenc}) you might as well set up the characters manually (which might be a bit of work):

\documentclass[]{article}

\usepackage[utf8]{inputenc}
\DeclareUnicodeCharacter{3C9}{\omega}% I don't want to do this for each greek character
\DeclareUnicodeCharacter{3B4}{\delta}% I don't want to do this for each greek character

\begin{document}
$ω$ is prettier than $δ$
\end{document}

Another (dirty) option using \usepackage[utf8x]{inputenc} would be

\documentclass[]{article}

\usepackage[utf8x]{inputenc}
\let\textomega\omega% dirty
\let\textdelta\delta% dirty

\begin{document}
$ω$ is prettier than $δ$
\end{document}