Number to words for NVDA

Author: Mateo Cedillo

Introduction:

I was inspired to make this little add-on, because some people use speech synthesizers like ETI-Eloquence which has some flaws when it comes to number processing, even if they are separated and confuses the user saying both numbers separated by spaces as if it were a decimal point.

This add-on improves the reading of numbers to words for these cases, it supports larger numbers and, in addition, the library supports many languages.

comparison of results between original and converted speech input

This comparison table demonstrates the differences between number processing on a synthesizer and num2words processing.

The following comparison has been evaluated using the IBMTTS driver for NVDA.

Using long numbers:

Language Original input Output Converted input
Spanish 921359131290481307233416326 nueve dos ún tres cinco nueve ún tres ún dos nueve cero cuatro ocho ún tres cero siete dos tres tres cuatro ún seis tres dos seis novecientos veintiuno cuatrillones trescientos cincuenta y nueve mil ciento treinta y uno trillones doscientos noventa mil cuatrocientos ochenta y uno billones trescientos siete mil doscientos treinta y tres millones cuatrocientos dieciséis mil trescientos veintiséis
English 921359131290481307233416326 nine two one three five nine one three one two nine zero four eight one three zero seven two three three four one six three two six nine hundred and twenty-one septillion, three hundred and fifty-nine sextillion, one hundred and thirty-one quintillion, two hundred and ninety quadrillion, four hundred and eighty-one trillion, three hundred and seven billion, two hundred and thirty-three million, four hundred and sixteen thousand, three hundred and twenty-six

Using spaces as separators (only in Spanish):

Usage:

This addon has three ways of using numbers to words:

Note: you can also copy the last result that was converted (see below)

Input gestures:

Important notes:

Compile this add-on:

Note: this add-on depends on a submodule, so:

  1. cd to this repo: cd num2words_nvda
  2. Type git submodule init and git submodule update in the console to clone the num2words library repository that is set as a module.
  3. If there are no errors, scons command should work correctly.

Add-on(s) I was inspired by

During the development of this add-on, I have been inspired by the following which I thank each of the authors of these add-ons:

Contact:

If you want to help improve this addon, you can send an email to angelitomateocedillo@gmail.com or make your contributions on the GitHub repo.


Changelog:

0.5.1

0.5

0.4.1

0.4

0.3

0.2

0.1