This Trick No Longer Works

Google has added a captcha to this service so it will no longer work.

OS X has the say command, which allows it to speak many different voices.  Linux users can also have this ability–although it is not as in-depth.

Installing Pre-requisite Software

The Raspberry Pi does not have mplayer  installed by default.  Fortunately, it can be installed via aptitude.

sudo apt-get update sudo apt-get install mplayer

Add Code to the .bashrc File

Add the following line to the bottom of the ~/.bashrc  file

function say { mplayer "http://translate.google.com/translate_tts?tl=en&q=$1"; }

Test the New Command

Save the file, exit, and log back in for the command to take effect.  Now, just type
say “whatever you want to say in quotes”  and the Pi will speak for you.
Depending what speaker setup you have, you may need to adjust some settings.  In particular, you can try telling the Pi what audio interface to use with the command

amixer cset numid=3 n
where n  is:
0=auto 1=analog 2=HDMI

Make the Say Command Scriptable By Adding it to /usr/bin

Create a new file: /usr/bin/say .  The file should read like this (slightly different that if it is in .bashrc )

#!/bin/bash mplayer "http://translate.google.com/translate_tts?tl=en&q=$1";

The file needs to be executable; there are two ways to do this:

chmod 755 /usr/bin/say

or

chmod +x /usr/bin/say