Text-to-Speech languages on K Touch

[diba]

Guys,
if you have on your K3 usertts and your respective languages installed, just copy the corresponding files from usertts/data to the K Touch. There you will find 3 different folders in the tts/data folder, common, high-quality, low-quality. Choose the files from the data folder of K3 and copy in the different folders, overwriting the respective files.

And your Touch is reading in your language. Easy this time.

diba

[Cokio]

But i don't have a K3...how can i get the tts/data folder?

[inameiname]

Originally Posted by Cokio

But i don't have a K3...how can i get the tts/data folder?

Here are links to two places where you can get the usertts/data:

http://gitbrew.org/~dasmoover/kindle/keyboard/usertts/


http://ge.tt/8HHrZe8

[ixtab]

Originally Posted by Cokio

But i don't have a K3...how can i get the tts/data folder?

You can look into this thread:
https://www.mobileread.com/forums/sho...d.php?t=153107

in the very first post, there is a link (to http://ge.tt/8HHrZe8 ) where you can download the relevant tts packs. What I did is download an appropriate source pack, as well as the usertts sources. In there, just take a look at the "usertts" executable and understand what it does.

Essentially, you have to rename the source files by replacing "#name#" by "tom" or "samantha", and "#number#" by "1" or "3" respectively. You will end up with a couple of files. On the KT, you will find (most of) these file names in the /mnt/us/tts/data/{common, low-quality, high-quality} folders, so just replace them with your new ones accordingly. It takes a bit of investigation and of copying around, but it works.

The above eas written from memory (don't have the device around), so I may have gotten some path names slightly wrong.

HTH

[Cokio]

I got it, thank you ixtab and inameiname.

[inameiname]

Originally Posted by Cokio

I got it, thank you ixtab and inameiname.

You're welcome.

Sadly, one of the voices I am still looking for is the British female voice. The only English one I could find besides my default American English ones is the British male one. So if anybody knows where I can get the female British voice, I'd love a heads up.

[ASnoopyDOG]

My new kindle touch tts look like this:



(german voice steffi - "kindle-voice-de-steffi.src")

But after restart I got an error:



Some idea?

Thanks in advance.

[ixtab]

I'm attaching both female and male german TTS packs (Steffi/Yannick) for the K-Touch. These work fine on my device.

Simply unzip into the tts folder, overwriting existing files. You can even completely delete the data folder and use the outcome of unzipping *both* files instead. (Still, it's always a good idea to have a backup of the original folder)

After you install them, you may need to "restart ttsd", "restart framework", or both. If you don't understand this, just restart the entire device ;-)

HTH

Update: Files removed from this post. See a few posts down to find updated and correctly working versions.

[ASnoopyDOG]

Thanks ixtab,
your files work for my when i ONLY replaced the files. But i got an error messages whenreplaced the folder with the other original files).
Greetings.

[ixtab]

Originally Posted by ASnoopyDOG

Thanks ixtab,
your files work for my when i ONLY replaced the files. But i got an error messages whenreplaced the folder with the other original files).
Greetings.

Sorry, didn't get that ;-)

Do you mean that applying only steffi.zip (or only yannick.zip) to a pre-existing "virgin" data folder didn't work, while removing the entire folder and then installing both files did? I did "patch" my installation with (only) the female german voice before, and it worked. That way, I could have german books read to me by the female voice, and english ones by the male voice. (It is much more hilarious the other way round though).

The entire TTS system is quite messy. For example I'm not quite sure even why the subfolders common, high- and low-quality exist in the first place. They don't seem to be referenced anywhere in the code, so I don't see a reason for them to exist at all. But then again: as long as it works, I don't see a reason to further dive into it.

I have also experienced some slight problems -- like the book occasionaly not starting to read with the male voice, but simply remaining quiet. Switching to female, turning TTS off then on, and switching to male again, worked though. As I said, it's a bit of a mess. Still, I never got error messages like the ones shown above. So if you find a reproducible way to obtain them, post some instructions...

[Mjamy]

Originally Posted by ixtab

I did "patch" my installation with (only) the female german voice before, and it worked. That way, I could have german books read to me by the female voice, and english ones by the male voice.

That sounds quite interesting! Can you specify what exactly you did to have german books read to you?

[ixtab]

Originally Posted by Mjamy

That sounds quite interesting! Can you specify what exactly you did to have german books read to you?

Ouch. I'm sure that I did it once, but simply unzipping only steffi.zip over the original tts folder probably will also result in an error.

AND NOW I'm screwed: I did make a backup of the partition, but only AFTER I already modified some of the tts files. So I'm now in the situation that I can't restore the factory-default anymore.

Would some kind soul who hasn't tampered with their tts folder (or who has a pristine backup) zip it up and send me a PM on where I can find it? Thanks a ton!

[diba]

You got a PM with the link.

I have done the same, Eng-male, Ger-female on my K3 weeks ago and Touch just when I got it. In the Ger loc of K3 I have even written this in the menu, so I change between Eng and Ger, not male-female.

[ixtab]

Muchas gracias, diba!

[ixtab]

Thanks to diba, I got everything together again. I'm currently uploading a file containing all 4 german/english voices, so you can combine them whichever way you want.

The upload will still take a few minutes; I'll post the link once it's done. If you can't wait, here is the README already:

You can mix and match these files as you please. (1 male, 1 female, of course):
Samantha/Tom = en/en
Samantha/Yannick = en/de
Steffi/Yannick = de/de
Steffi/Tom = de/en (*) see below

Simply copy the folders into tts/data, overwriting existing files.

NOTE: If you start from a pristine configuration AND want to keep Tom as
an english speaker, you will also need to copy the files for tom again. This
is because the configuration initially present on the device reuses some files
from the female configuration. So the combination Steffi/ Tom, if not
overwriting Tom's files, will have Tom speaking a rather interesting
mixture of english and german.

UPDATE: Here are the tts-* files: http://ge.tt/9XeUvcC