Why does the eSpeak pronunciation sound off even though I specify the language?

Question

I'm trying to use the command line version of eSpeak:

http://espeak.sourceforge.net/

It is possible to change the voices, but the language pronunciation remains American English.

For example:

espeak.exe -v es "como estas?"

Why does it sound like an American trying to speak Spanish, and not Spanish pronunciation?

Answer 1

Why does it sound like an American trying to speak Spanish, and not Spanish pronunciation?

Probably because that's how it's recorded in the voice file? Maybe they couldn't/didn't get a native speaker to do the job, hence the poor quality.

Here's what the espeak documentation has to say:

Help Needed

Many of these are just experimental attempts at these languages, produced after a quick reading of the corresponding article on wikipedia.org. They will need work or advice from native speakers to improve them. Please contact me if you want to advise or assist with these or other languages.

The sound of some phonemes may be poorly implemented, particularly [r] since I'm English and therefore unable to make a "proper" [r] sound.

A major factor is the rhythm or cadance. An Italian speaker told me the Italian voice improved from "difficult to understand" to "good" by changing the relative length of stressed syllables. Identifying unstressed function words in the xx_list file is also important to make the speech flow well. See Adding or Improving a Language.

Also,

fi Finnish This has had assistance from native speakers and should be usable.

fr French This has been improved by a native speaker, and should be OK.

So my guess above was correct. While the page does state that "Spanish has good spelling rules, so it should be OK", that should clearly indicates some amount of doubt.

Basically it's Free and Open Source, so you either take what you get, help improve it if you can, or go buy commercial software that can provide far better voice files created with the help of native speakers.

Answer 2

I think accent marks might be significant in the way eSpeak interprets the sounds, especially in Spanish. I've been studying this language lately using an online coaching tool, and I know it will complain when I leave out the acute accents.

Ergo, "como estas?" may sound significantly different from the proper "¿cómo estás?".

Try putting in the accent marks, and if it still sounds wrong, my hypothesis would be that the voice files aren't being properly read by eSpeak; maybe they weren't fully installed. However, if that were the case, you'd expect to see an error such as "Failed to read voice 'es'".

I hope this helps.