Taranis updated voice pack

User avatar
Kilrah
Posts: 11108
Joined: Sat Feb 18, 2012 6:56 pm
Country: Switzerland

Re: Taranis updated voice pack

Post by Kilrah »

Server address has changed. Just get it from within companion9x.

User avatar
Scott Page
Posts: 864
Joined: Wed Dec 28, 2011 3:32 am
Country: United States
Location: Tri-Cities, Washington State

Re: Taranis updated voice pack

Post by Scott Page »

mmatheny wrote:
Kilrah wrote:Just a little heads-up that I forgot to mention. Following the arrival of production radios I noticed the speaker was different than the one that was on our proto radios, and the sounds I had optimized for the proto sounded quite bad.

So I have tested a few different filters, and made a new pack that should sound lots better on the production radio (for English, that is). Of course FrSky have loaded the old one, so you may want to replace the audio files on the SD card by those available from c9x, or directly here:

http://85.18.253.250/voices/opentx-taranis/en/
Is this link up? I can't seem to get to it.

It's not working. I'm have posted it in a temporary location for you. https://dl.dropboxusercontent.com/u/178 ... Sounds.zip
mmatheny
Posts: 7
Joined: Fri Sep 06, 2013 7:29 pm
Country: -

Re: Taranis updated voice pack

Post by mmatheny »

How do I get it from within C9X? By downloading the full SD card?
User avatar
Kilrah
Posts: 11108
Joined: Sat Feb 18, 2012 6:56 pm
Country: Switzerland

Re: Taranis updated voice pack

Post by Kilrah »

Select the proper FW type, the voice language you want and click Download.
This will direct your web browser where the file is.
Attachments
28_c9x_prefs.png
sirzepp
Posts: 3
Joined: Wed Oct 30, 2013 1:56 pm
Country: -

Re: Taranis updated voice pack

Post by sirzepp »

Kilrah wrote:Select the proper FW type, the voice language you want and click Download.
This will direct your web browser where the file is.
.

I didn't know about that! Thanks.

Romushock
Posts: 47
Joined: Thu Nov 08, 2012 8:35 am
Country: -
Location: North Italy

Re: Taranis updated voice pack

Post by Romushock »

Hi, guys!
I've create a soundpack in russian language. Tested on Taranis and works great. Would like to ask developers to add an "ru" voice as optional to Companion preferences.
Thanks to all for amazing work :D

edit: added some model's names: copters, simulators, helis, planes. Not so much - but better that nothing...
Attachments
ru+models.rar
(11.36 MiB) Downloaded 592 times
ru.zip
(13.64 MiB) Downloaded 807 times
Emanuel
Posts: 11
Joined: Sun Dec 01, 2013 3:23 pm
Country: -

Re: Taranis updated voice pack

Post by Emanuel »

Hi,

I am new to Taranis and this forum. Its been very helpfull so far. With the help of this thread I managed to make my own set of sound files. I wanted a voice that sounded a little bit friendlier and human. I found a voice I liked on the Acapella demo website.
http://www.acapela-group.com/text-to-sp ... -demo.html
The Austalian English voice Lisa sounded nice so I made a complete set of sounds using that voice.
I don't know how to make an attachment for this forum so uploaded the file to wetransfer. If you try the files and like them, maybe someone with more experience on this forum can make it more permanently available.
Here is the link: https://www.wetransfer.com/downloads/12 ... 037/6311a1
Its only active for 7 days.

If you find any errors, please let me know so I can correct the mistake.

Thanks for all the support!!
User avatar
Scott Page
Posts: 864
Joined: Wed Dec 28, 2011 3:32 am
Country: United States
Location: Tri-Cities, Washington State

Re: Taranis updated voice pack

Post by Scott Page »

Emanuel wrote:Hi,

I am new to Taranis and this forum. Its been very helpfull so far. With the help of this thread I managed to make my own set of sound files. I wanted a voice that sounded a little bit friendlier and human. I found a voice I liked on the Acapella demo website.
http://www.acapela-group.com/text-to-sp ... -demo.html
The Austalian English voice Lisa sounded nice so I made a complete set of sounds using that voice.
I don't know how to make an attachment for this forum so uploaded the file to wetransfer. If you try the files and like them, maybe someone with more experience on this forum can make it more permanently available.
Here is the link: https://www.wetransfer.com/downloads/12 ... 037/6311a1
Its only active for 7 days.
With Emanuel's permission I have given his voice project a (more or less) permanent home @ https://drive.google.com/file/d/0B5COff ... sp=sharing
staxin
Posts: 3
Joined: Thu Dec 05, 2013 5:47 am
Country: Denmark
Location: Bronshoj, Copenhagen

Re: Taranis updated voice pack

Post by staxin »

Emanuel wrote:Hi,

I am new to Taranis and this forum. Its been very helpfull so far. With the help of this thread I managed to make my own set of sound files. I wanted a voice that sounded a little bit friendlier and human. I found a voice I liked on the Acapella demo website.
http://www.acapela-group.com/text-to-sp ... -demo.html
The Austalian English voice Lisa sounded nice so I made a complete set of sounds using that voice.
I don't know how to make an attachment for this forum so uploaded the file to wetransfer. If you try the files and like them, maybe someone with more experience on this forum can make it more permanently available.
Here is the link: https://www.wetransfer.com/downloads/12 ... 037/6311a1
Its only active for 7 days.

If you find any errors, please let me know so I can correct the mistake.

Thanks for all the support!!
I also think Lisa sounds nice :P , so I have adopted your work and now want to add a few extra sounds regarding my FPV camera gimbal: "Camera tilt down", "Camera tilt front" and "Camera tilt manual" and there might be more to add in the future....
I then went to acapela-box.com and found "English (AU) Lisa", however, she doesn't sound exactly like your Lisa :(
So question is if you post-processed the voice in any way? If yes how did you do it?

Thanks so far,
Steffen
-----
DjiPhantom w. 3S@4400mAh, GoPro3/Tarot Gimbal, Fully 5.8Ghz FPV'ed + OSD
FrSKY Taranis X9D + X8R, 7" Monitor
velzemans
Posts: 6
Joined: Tue Oct 22, 2013 9:07 am
Country: -

Re: Taranis dutch voice pack

Post by velzemans »

I created a Dutch voicepack for the Taranis, it is available for download HERE
staxin
Posts: 3
Joined: Thu Dec 05, 2013 5:47 am
Country: Denmark
Location: Bronshoj, Copenhagen

Re: Taranis updated voice pack

Post by staxin »

staxin wrote: I also think Lisa sounds nice :P , so I have adopted your work and now want to add a few extra sounds regarding my FPV camera gimbal: "Camera tilt down", "Camera tilt front" and "Camera tilt manual" and there might be more to add in the future....
I then went to acapela-box.com and found "English (AU) Lisa", however, she doesn't sound exactly like your Lisa :(
So question is if you post-processed the voice in any way? If yes how did you do it?

Thanks so far,
Steffen
Never mind, I found an even better voice, and re-processed all the commands I use ;)
But thanks anyway for pointing to acapela-box.com
-----
DjiPhantom w. 3S@4400mAh, GoPro3/Tarot Gimbal, Fully 5.8Ghz FPV'ed + OSD
FrSKY Taranis X9D + X8R, 7" Monitor
PeterV
Posts: 76
Joined: Sun Feb 26, 2012 9:33 pm
Country: -

Re: Taranis updated voice pack

Post by PeterV »

How did you record the results from Acapela ?
thanks.

EDIT: I found my download manager could capture the current played sound. As an MP3... then I go and convert it to Wav.
staxin
Posts: 3
Joined: Thu Dec 05, 2013 5:47 am
Country: Denmark
Location: Bronshoj, Copenhagen

Re: Taranis updated voice pack

Post by staxin »

there is a good how-to here: http://www.youtube.com/watch?v=pi_KAbgJ3vE
-----
DjiPhantom w. 3S@4400mAh, GoPro3/Tarot Gimbal, Fully 5.8Ghz FPV'ed + OSD
FrSKY Taranis X9D + X8R, 7" Monitor
User avatar
jhsa
Posts: 19480
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Taranis updated voice pack

Post by jhsa »

I tried to do as you explained on the video but I couldn't find the mp3 file. So here is what I've done.

search for a pulgin for firefox (don't know if it exists for other browsers) called "UnPlug". install it.
write your text on the acapela demo page. and click listen so the site loads your text. after, right click somewhere near the play button and choose "UnPlug". this will open a download window. click on the arrow near the download button and choose "open in a new tab". the file will start playing in a new tab. right click and save as mp3. change the name to whatever you want and save it.
I tried to save directly by hitting download but it didn't work.. had to open it in a new tab as explained..
I hope it helps..

João
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW
Hawk600
Posts: 22
Joined: Tue Jan 07, 2014 4:58 am
Country: -

Re: Taranis updated voice pack

Post by Hawk600 »

Just a feedback that "UnPlug" plugin for Firefox works great. You have to play the acapella demo with the text you want and after that just click on the UnPlug icon and it will open a list of files. The one you looking will be the last one down the bottom of the list. Right click on it, choose open in new tab and as soon as the new tab opens you can save the audio as mp3.
Thanks for the tip João!
User avatar
jlummel
Posts: 53
Joined: Sat Dec 15, 2012 5:56 pm
Country: -
Location: Windy Northern Arizona

Re: Taranis updated voice pack

Post by jlummel »

When I updated some of the phrases in the en_ivo_emma sound pack (see forum message Taranis voices), I used the webpage demo at http://www.ivona.com and entered several phrases at a time with a comma in between each phrase. When I played the phrases back I hooked a Tascam DR-5 digital recorder to the speaker out of the computer and recorded the output (though almost any recording device should work such as an iPhone or an Android phone/tablet, etc).

I also modified the phrases phonetically to produce a pleasing result (this changes based on the voice used, BTW), see http://www.ivona.com/us/?tk=hTBKfXOBIw for examples.

I then took the entire wave file into Audacity, converted to mono, normalized to -3db and set the project format to 32kHz. I then went thru and highlighted each phrase individually and used "Export Selection" to create the individual wave files.

You would have to do several phrases at a time in batches as the text box on that website only allows 255 characters, but even batch processing this way can be done at a reasonable clip. This should also be a faster way to create sound files using other TTS applications and websites.

James

PS- you could also use the currently free Ivona beta on an Android phone or tablet and then record to another phone or tablet (or a digital recorder, if you have one), to use the Ivona TTS you don't necessarily need to use the website or download the trial on your computer...
DarKHawK
Posts: 29
Joined: Thu Apr 04, 2013 12:36 pm
Country: -

Re: Taranis updated voice pack

Post by DarKHawK »

If you can can you update the txt files in https://www.dropbox.com/s/zpu6roejenkpn ... txt_en.zip

I will then reprocess all the voice packs and zip and upload again.

The software I use is

1. text2speak engines
2. Balabolka, text to speech software, has a batch mode for txt files, aka it processes the txt files in zip and creates wav files. (its free)
3. r8brain, has a batch mode to resample wav files. (free version works fine)

step 2 takes about a minute to run and step 3 takes about 2 minutes then compressing a few seconds.




jlummel wrote:When I updated some of the phrases in the en_ivo_emma sound pack (see forum message Taranis voices), I used the webpage demo at http://www.ivona.com and entered several phrases at a time with a comma in between each phrase. When I played the phrases back I hooked a Tascam DR-5 digital recorder to the speaker out of the computer and recorded the output (though almost any recording device should work such as an iPhone or an Android phone/tablet, etc).

I also modified the phrases phonetically to produce a pleasing result (this changes based on the voice used, BTW), see http://www.ivona.com/us/?tk=hTBKfXOBIw for examples.

I then took the entire wave file into Audacity, converted to mono, normalized to -3db and set the project format to 32kHz. I then went thru and highlighted each phrase individually and used "Export Selection" to create the individual wave files.

You would have to do several phrases at a time in batches as the text box on that website only allows 255 characters, but even batch processing this way can be done at a reasonable clip. This should also be a faster way to create sound files using other TTS applications and websites.

James

PS- you could also use the currently free Ivona beta on an Android phone or tablet and then record to another phone or tablet (or a digital recorder, if you have one), to use the Ivona TTS you don't necessarily need to use the website or download the trial on your computer...


Sent from my K00C using Tapatalk
[url=https://www.viagrasansordonnancefr.com/sildenafil-100-milligrams-ligne/]sildenafil 100 milligrams ligne www.viagrasansordonnancefr.com[/url]
User avatar
jlummel
Posts: 53
Joined: Sat Dec 15, 2012 5:56 pm
Country: -
Location: Windy Northern Arizona

Re: Taranis updated voice pack

Post by jlummel »

DarKHawK wrote:If you can can you update the txt files in https://www.dropbox.com/s/zpu6roejenkpn ... txt_en.zip

I will then reprocess all the voice packs and zip and upload again.
...
The only problem I found with the text files in that archive was the radio battery low warning, the 'y' was left off of battery. And I prefer "Throttle Warning" rather than "Throttle Alert".

Each TTS program and even the voices in the same TTS program has different problems with how it pronounces words. An example of this is the word "watt", or "watts". Inova Emma thinks it's trying to say "What?" and then says it with attitude. I had to fool it into pronouncing the word normally using "1 watt" or "88 watts" as the phrase and then cutting out the "1" or the "88" I didn't need. The number "3" was also a challenge and I had to fool it into pronouncing it correctly as well, as well as "G" and "Gs". They all required editing to get it right, usually to remove the extra words necessary to get a correct pronunciation.

Batch processing can and does work for the majority of phrases, but you need to manually go thru all the phrases and then tweak the ones that didn't come out right. But these tweaks are not guaranteed to work with any other TTS engine, nor even with any other voice in the same TTS engine. But once you get the phrase tweaked just right, it should always work for that voice on that TTS engine it was tweaked for.

For example, go to http://www.ivona.com/us/?tk=hTBKfXOBIw and select other voices and play it back. Not all voices work with the particular tweaks I made to the phrases on that list. But with minor changes, you can get all voices to properly speak these phrases. Try Jennifer and listen to how she says the tweaked "Taranis" as an example, or Gwyneth and the tweaked word "GPS" or the tweaked word "RF".

James
User avatar
maccross
Posts: 11
Joined: Sat Jun 08, 2013 4:37 am
Country: -

Re: Taranis updated voice pack

Post by maccross »

Sorry if I'm bringing a possible issue in the wrong place.
I'm trying to make a portuguese (brazilian accent) version of the present voice pack taking the english version as a guide.
I started with the numbers because my first interest is altitude info.

One problem is that in portuguese we say:
example : 25
twenty AND five instead of twenty five
(actually this is no problem because the sound file is unique)

but it sounds weird when it comes to hundreds
in portuguese it goes like this:
example: 256
two hundred AND fifty AND six (here the fifty AND six is solved because of the unique file but the two hundred AND will be weird when the number is 200 (no AND in this case!)
The same applies to thousands.

Of course I can live with this (and many portuguese speaking fliers too) but it would be much more elegant to have the numbers spoken the correct way

All comments will be apreciated. And a fix even more.
User avatar
jhsa
Posts: 19480
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Taranis updated voice pack

Post by jhsa »

Well in Portugal we also say "Duzentos E cinquenta e seis" for 256. For us that is correct ;)

Joao

Sent from my GT-I9195 using Tapatalk
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW
User avatar
maccross
Posts: 11
Joined: Sat Jun 08, 2013 4:37 am
Country: -

Re: Taranis updated voice pack

Post by maccross »

João
jhsa wrote:Well in Portugal we also say "Duzentos E cinquenta e seis" for 256. For us that is correct ;)Joao
yha. I thought you would confirm what I was saying, glad to know they still speak the same language ;)
User avatar
jhsa
Posts: 19480
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Taranis updated voice pack

Post by jhsa »

Yeah, we Portuguese people still speak Portuguese. For how long, I don't know.. :mrgreen:

João
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW
User avatar
maccross
Posts: 11
Joined: Sat Jun 08, 2013 4:37 am
Country: -

Re: Taranis updated voice pack

Post by maccross »

This is what I did using this recipe https://www.youtube.com/watch?v=pi_KAbgJ3vE
At least jhsa can comment ;)
Cheers
Attachments
br.zip
Portuguese (Brazil) system voice pack
(8.81 MiB) Downloaded 439 times
User avatar
jhsa
Posts: 19480
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Taranis updated voice pack

Post by jhsa »

Is that the video you wanted to post? Because that is the video you posted before..

Joao

Sent from my GT-I9195 using Tapatalk
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW
User avatar
maccross
Posts: 11
Joined: Sat Jun 08, 2013 4:37 am
Country: -

Re: Taranis updated voice pack

Post by maccross »

João

Yes it is.
It seems you are having trouble making it work.
May be this can help you:
hack1.jpg
hack2.jpg
hack3.jpg
hack4.jpg
hack5.jpg
User avatar
jhsa
Posts: 19480
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Taranis updated voice pack

Post by jhsa »

Ahh, ok. Thanks. I did have trouble with the page source code. But now I found the "UnPlug" plugin for firefox and it does all that work for me. Just right click, choose UnPlug and save the file as mp3. Very simple. Try it ;)

Joao

Sent from my GT-I9195 using Tapatalk
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW
User avatar
maccross
Posts: 11
Joined: Sat Jun 08, 2013 4:37 am
Country: -

Re: Taranis updated voice pack

Post by maccross »

Glad you found a way arround, João.

Now back to the main subject.

Hi software masters.
I found some portuguese Taranis voices using the companion9x here: http://downloads.open-tx.org/voices/opentx-taranis/pt/.
They are arranged somewhat different from the english version.
Should I make my brazilian-portuguese version accordingly or are those portuguese versions outdated (file date is May 31,2013)?
User avatar
jlummel
Posts: 53
Joined: Sat Dec 15, 2012 5:56 pm
Country: -
Location: Windy Northern Arizona

Re: Taranis updated voice pack

Post by jlummel »

DarKHawK wrote:If you can can you update the txt files in https://www.dropbox.com/s/zpu6roejenkpn ... txt_en.zip

I will then reprocess all the voice packs and zip and upload again.
...
Found one more problem with the list of text files you mention. 110 should be deleted, it currently says 'volt'.

I ran into a problem when I had it reading out my throttle timer when I hit a switch, it would say "4 minutes volt 33 seconds" and I traced down the error to file 0110.wav. This file is not among the files in the stock sound pack and looks like if may have said "colon" at some point in time rather than "volt" (which would have been redundant, btw, as 0115.wav says 'volt').

Anyone using the INVONA Emma sound pack should delete this wav file also.

James

PS- you could also replace 0110.wav with 'and' so it would read 'and' every where there is a colon and for timers it would be "4 minutes and 33 seconds", which would sound nice too
Attachments
Ivona_Emma_0110.zip
This is 0110.wav with Ivona Emma saying 'and', otherwise you should delete the existing 0110.wav
(24.63 KiB) Downloaded 435 times
User avatar
jlummel
Posts: 53
Joined: Sat Dec 15, 2012 5:56 pm
Country: -
Location: Windy Northern Arizona

Re: Taranis updated voice pack

Post by jlummel »

Here are some optional additions to the IVONA Emma sound pack. There are two archives, one is aircraft names that are spoken when the model is selected. The other has two sound files for indicating whether the model uses an internal RF module or an external RF module (and is module agnostic) and can be played after the model name when the model is selected. You simply unpack these onto the SD card into the \SOUNDS\EN\ directory to use (I've standardized on using Caps for aircraft names only and lower case on all other sound files on my Taranis).

Here are the planes listed in the zip file aircraft names:
AeroSky 550 Hexacopter
1/4th-Scale J3 Piper Cub
1/6th-Scale Super Cub - EP
Dynam Spitfire foamy
F4F Funfighter foamy
Fong-3D
Hobby King Super Cub - EP
HobbiStar Mark 3
Kadet Senior
NitroPlanes Low Wing 60
LT-40 Kadet
Micro Albatros DV foamy
Micro Cessna 182 foamy
Millennium Master
Mini Pitts
Neptune EDF foamy
Freewing Pandora foamy
Sky Trainer 400 foamy
A10 Warthog EDF

Hope you like!

James
Attachments
Model Names.zip
IVONA Emma aircraft model name wave files
(1.92 MiB) Downloaded 445 times
RF Module.zip
IVONA Emma RF module wave files
(136.93 KiB) Downloaded 434 times
User avatar
xxdvdxx
Posts: 18
Joined: Mon Jan 13, 2014 9:51 am
Country: Italy

Re: Taranis updated voice pack

Post by xxdvdxx »

The download link doesn't work
__________________________Equipment_______________________________
TX: X9DP(B01 ver),Fw:OTx v2.2.0 EEPR 218,FwRadio:(LBT151223)
RXs: FrSky X8R(LBT151118), FrSky X4RSB(LBT170322), FrSky D4r2(LBT151118)
Mods:extAnt,6rsw,M9gimb
_____________________________

Post Reply

Return to “openTx for FrSky radios”