How to make Your own sound files to Jeti?
In basic it’s easy. Drop a file to Jeti and select it. In practice there’s a few pointers on should take in to consideration.
First of all, not any soundfile works. Second of all, it Would be cool if all the sounds would sound pretty much the same, this is of course a matter of opinion…
Update: You can also get your sound-files from RC-Thoughts! Just go here to get the files! Use this guide to do editing if You need to.
The original sounds (speech) in Jeti is from Ivona Text-To-Speech, You can use their webpage to generate Your own sounds that sound exactly like Jeti’s original. (The speech profile in english is “English, American, Jennifer”). The Ivona page is http://www.ivona.com/en/.
Since AT&T Text-To-Speech is now down (forever apparently) this guide is now updated to use Ivona.
Unfortunately Ivona does not allow us to save the spoken text to file. So we need to record it with Audacity. So. let’s download Audacity, a free and very good sound-editor from http://audacityteam.org/ (do NOT download anymore from sourgeforce!) and install it.
After starting Audacity it looks like this:
First we need to make the Audacity to record the voice played by Ivona. Select your recording input as “Microsoft Sound Mapper – Input” as in picture:
Note: Set recording volume as low as possible = 0.00:
Now go to Ivona and play with commas, exclamation mark, capital letters and so on to get the pronunciation to what you want. when you’re satisfied leave the settings as they are and switch to audacity and start recording by pressing the record-button:
Jump back to your browser and Ivona, press the “Play”-button:
Go back to audacity and immediately press the “Stop”-button:
Now Audacity should something like this:
You’ll see there’s extra space (silence) in the beginning and in the end of file. Also the dynamic (“volumerange”) is pretty low, it’s not reaching to 1.0 on the vertical axis. But let’s fix the silence first.
Select with mouse the area you wan’t to remove, click on the center-line and drag the mouse until You have selected the area:
Simply press delete (DEL) button. Do the same to the end.
After that it’s time to maximize the dynamic range (volume) of the sound-file, let do something called normalization. First press CTRL-A to select the whole clip:
Then in the menu go to “Effect” -> “Normalize”:
The default maximum amplitude is -1.0 dB, I usually change this to -0,5 dB. Press the preview, if there’s any distortion at all lower the maximum amplitude to for example -1,5 dB and preview again. After it sounds good press ok.
Your file is now normalized to a larger dynamic area, it sound’s “louder”, this can also be seen in the waveform, it’s “taller”:
Now we need to save the file, go to “File” -> “Export audio”, select the file type to “WAV (Microsoft) Signed 16 bit PCM” and click save.
This is important! Jeti needs (want’s) the metadata to be stripped! So make sure there is nothing in the following screen before clicking the ok-button!
So, now You have the audio-file on Your computer. Move it to Your Jeti the usual way. Drop the file to right directory:
And use it where ever You wan’t to :)
Sometimes it just is easy :)
Hi
The link to the AT&T text to speech doesn't work?
Updated the guide to use Ivona. Thanks for heads up!
Great article as always, thanks!
Quick question.
What went wrong with sourceforge? Curious why you recommend not using that site?
Because of this: http://www.howtogeek.com/218764/warning-don’t-download-software-from-sourceforge-if-you-can-help-it/ as it says in the guide :)
I have set up audacity as you describe, but it wont record from Ivona unless I turn the record volume up. It seems to be using the PC mic, even though I have it selected to use as you describe. Any tips?
You need to enable stereo mix, look here: http://www.howtogeek.com/217348/how-to-record-the-sound-coming-from-your-pc-even-without-stereo-mix/
Or better, use my Text-To-Speech service, look for the TTS up in the menu, you can download ready wav's :)
Sad to hear that AT&T Text-To-Speech is now down. It was so simple tu use. No need any complex download or user app…. also was MAC friendly…. I hope in the future someone come out with something similar.
Have a look in the menu. (Hint: TTS) :)
Many, many, many, many THANKS!!!! ;)))
Hi from Ireland…im a new DS 16 user and have a quick question please….i loaded on my custom audio .wav files and they all work fine however as i have a lot of them im wondering is there a quick way / shortcut to navigate between the pages without having to scroll to the bottom of each page and then select NEXT>>
eg , if i need to select " Throttle Cur " i have to scroll to the near the last page to find it…any help is appreciated.
When you enter the file-page scroll one up, then you are at the bottom and can quickly press next to get to following page. As far as I know it's the quickest way to jump pages.
thanks tero..
one last query….when adding a custom wav file to the SD file structure, should i use the < AUDIO> or
< VOICE > folder ?
By the way , you are a great resource for the JETI user and thanks for all you do. Im learning a lot from reading your blog. top job sir
They go to Audio-folder. Enjoy your Jeti, lot to learn, many times there's many ways to do one thing :)
Tero – looks like Ivona is now no more and Amazon want your money for their replacement. I've used the tool you have here to create the sounds I need, but a bit more control would be handy.
Thanks for your efforts!
What control do you need?
Amazon Web Services "Polly" service is great for generating PCM or MP3 files from text which you can then convert to .wav for Jeti. You can even script it using the aws command line:
aws polly synthesize-speech –output-format "mp3" –voice-id "Joanna" –text "Stall turn, quarter rolls up and down" stallTurnCenter.mp3
Yeah. But have you had a look on their API? :)
I use many wav files for my own radio.
A little script generates all the files . I use pico2wave and sox, this is running in linux or a raspberry.
It is followed with one line for each wav file (I use nearly 200 sound files !)
w "file_name" "text …."
Robert
#!/bin/bash# ce script crée de nombreux fichiers "son " à partir d'un texte
# utilise pico2wave
#il écrit le fichier .wav dans le folder DESTINATION
#DESTINATION="/mnt/data/d/Develop/RCCL_V5/SND/"
DESTINATION="/media/robert/RCCL_V5B/SND/"
# fonction "w" pour créer un fichier wav
# 1er paramètre = nom du fichier (sans .wav et sans espace)
# 2ème paramètre = le texte
function w
{
pico2wave --wave=$DESTINATION$1_o.wav --lang=fr-FR """ $2 """
sox $DESTINATION$1_o.wav $DESTINATION$1_r.wav reverse
sox $DESTINATION$1_r.wav $DESTINATION$1_rs.wav silence 1 0.200 0
sox $DESTINATION$1_rs.wav $DESTINATION$1.WAV reverse
rm $DESTINATION$1_*.*
echo $1 $2
}
#remarque sox est utilisé pour supprimer le silence à la fin
#set -x # pour debug set +x pour reset
#
w "GEASW15" "Planeur A S W quinze électrique"
w "GEGRAPH2" "Graphite deux électrique"
Just download free program Balabolka, you can type in your speech and save to a wave file on your computer, dead easy.
or, Google text to speech is very easy to use and can also save files as wav
Hello Tero,
This is Davy from Ecuador, South America
Your help with Jeti is invaluable and I greatly appreciate that
Excuse my lack of ability in Jeti because I am just starting new with a DS12 deluxe package.
I just got done with your great TTS tool to have in audio my f3a pattern schedules. My objective is to fly two different maneuver schedules told by audio to me from my transmitter. How can I do this?
I have seen (on Harry Curzon video) how it is set up with a switch on the transmitter to listen to the next maneuver in the audio file. How can I program to have two files? For example first flight toggle switch flies schedule P and on another flight schedule F
That is simple, use another switch as multiplier. If Switch 1 is ON then Switch 2 always goes to schedule P, if Switch 1 is ON then Switch 2 goes to schedule F. Use logical switches for launching the events.