GPS Centar
GPS Centar
GPS Centar forum forum forum
GPS navigacioni urejdaji, oprema, programi, saveti i sve ono sto mislite da ima veze sa GPS-om : FORUM :
 FAQ - Često Postavljana PitanjaFAQ - Često Postavljana Pitanja   TražiTraži   Lista članovaLista članova   Korisničke grupeKorisničke grupe   Registruj seRegistruj se 
 ProfilProfil   Proveri privatne porukeProveri privatne poruke   PristupiPristupi 

Uputstvo za pravljenje glasova za TomTom

Napiši novu temu   Odgovori na poruku forum forum -> GPS Software
Pogledaj prethodnu temu :: Pogledaj sledeću temu  
Autor Poruka

Pridružio: 22 Feb 2007
Poruke: 1495
Lokacija: N_44.84080 E_20.39600

PorukaPoslao: Sre Mar 21, 2007 12:39 am    Naslov: Uputstvo za pravljenje glasova za TomTom Odgovoriti sa citatom

U sred pisanja mog uputstva, shvatio sam da je tesko opisati jednostavno nacin rada na koji ja to radim, i posle malo googlanja naisao sam na jedno slicno malo vise user friendly resenje koje prenosimo u celosti uz nase komentare i sugestije. Uputstvo je dugacko zato sto je do najsitnijeg detalja sve opisano - uopst nije niti tesko niti komplikovano

Uputstvo je preuzeto sa sajta:
(Za sada na engleskom)

Moje komentare i sugestije videcete podebljano

Step 1: Preliminary

In this section I will describe a little about how TomTom voices work, as well as explain what software you will need. Basically this step will get you to a stage where your ready to hit record!

a) Get to know how it all works.

I think to start it's important, especially if your new to this, to explain how this all works, so here it goes:

A voice profile is created in three steps, firstly the 59 individual voice commands are recorded ("Go straight on", "After", "Keep Left" etc.), then the standard sound format is encoded into the OGG Vorbis format used by TomTom, and lastly all the individually encoded files get compacted into two files specially named so that TomTom can use it.

Each voice profile contains either two or three files; the voice data (.chk), the index file ( .vif), and possibly the image ( .bmp, TomTom5+ only).

In order for TomTom to recognise your voice, it must be properly named in the format "data[number].chk". Each official voice released by TomTom has a unique file number, for example data04.chk is Tim UK, and data15.chk is Eva - Nederland's. As far as I am aware you cannot change this, however data61.chk to data99.chk (TomTom 5) and data71.chk to data99.chk (Pre TomTom 5), are recognised as 'Special Voices' by TomTom, I would always recommend using the special range instead of overwriting the official ones to avoid confusion.

If your using TomTom5, the voice files are stored in the folder 'Voices' under the tomtom directory (the location of this folder depends on your device and where you installed tomtom in the first place!), if your not using TomTom 5, get your credit card out and buy it now, your missing out! Of course until then, you can find the voices in the 'TomTom' folder. I'm fairly sure this is the same whatever type of TomTom you are using, but please correct me if I'm wrong!

OK, hopefully by now you have a rough idea, now read on to learn how to create your own voices.


b) Get and install the software, and create the folder structure.

For this guide, create a folder 'Tomtom voices' in the root of your 'C Drive'. To do this navigate to the 'C Drive' through 'My Computer', double click on the 'C Drive', then expand the 'file' menu, then choose 'new' > 'folder'.

For this guide to work, you will also need three crucial bits of software; Audacity, win:viftool2 (created by ME), plus my template OGG files (this will save you the hassle of naming each file).

Audacity (2.31 mb) -

Author: Audacity Developer Team

- I have chosen audacity purely and simply because I have been using it for quite a while, and I still can't get over how absolutely FANTASTIC it is, oh yeah and did I mention it's FREE? This is a great piece of software for any use, however the three major benefits of using audacity to create TomTom voices are:

1. You can create all of your 59 voice file in the same project window quickly and easily, then you can tinker with all your files at the same time.
2. It has a fantastic tool that lets you remove background noises and any static you might have, particularly if you are using a cheap microphone, full credit to Dominic Mazzoni for creating this tool!
3. It can export the file directly to the OGG Vorbis format that TomTom uses, saving you so much time!

Install Audacity but do not change any of the default settings just yet, I'll tell you exactly what to change later.

win:viftool2 (2315kb) -

Author: Graeme Lucas (That's Me!)

- win:viftool2 is a windows GUI Wizard application that I wrote in VB. win:viftool2 makes the final section far more easy than traditional methods as it avoids the need of using the command prompt. Download the setup program using the above link, and then install it using the included wizard. The program MUST be installed to c:\winviftool, so please DO NOT MOVE IT once it's installed otherwise it may crash!

Template OGG File Names (32kb) -

Author: Graeme Lucas (That's Me!)

- Download and extract the files into a folder named 'templates' under the 'Tomtom voices' folder you just created (C:\TomTom Voices\Templates). This archive contains the 59 ready named OGG files, all empty. This is not an essential step but will save you time when it comes to saving the files later in step 3!

Step 2: Record & Master

Now all the boring bits are done, your ready to go, again the following steps have screen shot, so be patient as they may take a while to load!

Some tips before you start:
1)Preuzmite program za raspakivanje TomTom vif fajlova tj glasova
2)Napravite u My Documents direktorijum "glas"
3)Raspakujte u taj direktorijum viftool
4)Nadjite medju vasim postojecim glasovima fajl koji se zove data47.vif i data47.chk i presnimite ih u direktorijum "glas"
5)Startujte command promt: Start>All Programs>Accessories>Command Prompt u kucajte dolenavedeno (na kraju svakog reda pritisnite enter)

cd glas
viftool split data47.vif

10)Sada mozete zatvoriti command prompt

U direktorijmu "glas" sada se nalazi raspakovano svih 59 reci / recenica na srpskom jeziku koje je potrebno procitati - mozete ih ucitati u gorespomenuti Audacity i cuti sta tacno treba izgovoriti.

Spisak potrebnih fajlova (ne texta potrebnog za citanje)je:

* Do not attempt to do all 59 samples straight away, do some tests with some of the longer commands, like "You have arrived at your destination", then play it back, it will give a good representation as to if you've got the right volume setting, as this will give you the opportunity to raise or lower the input volume beforehand, or move the voice-ee. The blue wave representation marks should reach the limits of the preview window, but not exceed it. Too quiet and it won't be loud enough, too loud and it will cause clipping (distortion), here is some examples of what the blue marks should look like:

too quiet >

too loud >

ahhh, just right! >

* Try and leave a silence gap before and after each sample, to avoid accidentally cutting off part of the sample, we will cut the sample down later on.
* Always read the samples in order of my recommendation text file, as this will help later on.
1.) Cue the voice-ee, hit record ( ) and starting from the top, cue the voice-ee to speak each command, hitting stop after each sample. When you hit the record button, Audacity will automatically start on a new sample window, which makes things quicker and easier! When they complete each line, hit the stop button
2.) Repeat step 1 for each of the 59 voice commands, ensuring you leave a small gap before and after each clip, they should look something like this (This was an example of "You have reached your destination)
2.) Repeat step 1 for each of the 59 voice commands, ensuring you leave a small gap before and after each clip, they should look something like this (This was an example of "You have reached your destination)

3.) Once that is complete, it should look something like this.

TIP: You may like to play back some of the clips, to do this you must click 'solo' on the sample you wish to play back and then hit play, otherwise they will ALL PLAY together and that's just scary, and loud! Use the scroll bar on the right hand side to view all the samples.

4.) Before you do anything else record one more sample, about 5 seconds of nothing but any background noise and static that may be around, this sample will be used to profile and remove the background noise from your other clips, neato!
5.) Next you will profile the background noise. To do this click on the information pane (Where it says 'Mono, 22050Hz etc.) of the last 'background noise' sample you just made, or by clicking and drag-selecting it using the 'selection tool'

6.) Select Effect > Noise Removal... and click on the 'Get noise profile' button, this will analyze the sample, and it will happen very quickly, so don't think it hasn't worked, it will automatically close the 'Noise Removal..' window and return you to the main screen.

TIP: do not remove the 'background noise' sample just yet!

7.) Now you can apply the noise removal to all you samples. Do this by pressing CTRL+A, you will notice this will highlight all the samples. Then go back into 'Noise Removal..' and hit the 'Remove Noise' button.

TIP: you may find you want to move the less/more slider. However be careful, too less and you'll still here background noise, too much and the voice will sound 'muffled'.

8.) Once the task has finished, you will noticed the 'gaps' before and after the sample will now look more like a straight line than jagged one, this will indicate the 'noise removal' has worked. Now you can delete the 'background noise' sample by clicking on the 'X' on the top left hand side of the sample window.
9.) Next you will cut down each clip by removing the unwanted gaps before and after each sample.

This will probably be the most difficult bit to get right, as it's important to leave just enough gap after each sample to that it will sound natural when on the road.

It you leave too much gap it will sounds like the person is having a nap between each part of the command (After.............100...............yards...............keep left!). On the flipside if you cut too much, it will all sound like a racetrack (After100yardskeepleft!).

I achieved this my clipping as closely before the clip as possible, and leaving an specific amount afterwards. On top of this, it's important to remember certain samples will require more gaps afterwards than another. For example, the sample "After" will require more gap afterwards than "100" as in normal speech you would pause, for example "After, 200 yards". Below is a screenshot of about how much I recommend should be left. But this is a rough guide.

Once you repeating step 9 for all 59 samples, you have completed this section of the guide. Now you are ready to convert and save each sample. You should now have a list of 59 samples in order of how they are listed on the guide sheet, they should all be free of any background noise, and should be cut so that there are no big gaps are before or after each sample.

Step 3: Save & Export

Once you have recorded and mastered each of the 59 clips, you are now ready to save and export your samples into the OGG Vorbis format.

Before you continue, you should create a copy of the 'Template' folder (That by now should contain the 59 empty OGG files).

Make a copy of the 'Template' folder and rename it to a name that will help you recognise the voice (eg The Name of the Voice-ee.) for this guide I have called the folder 'Graeme' as it will contain the voice of me! It should look something like this.

The reason why you created the 'Template' folder was so that you have a central copy just in case something goes wrong and you need to start over, then you can delete the entire folder (eg 'Graeme') and then re-copy the 'Template' folder.

Now you are ready to save.
1.) Select the first sample by clicking on the on the information pane (Where it says Mono, 22050Hz etc.), this will select the entire sample (You must select the sample before continuing).

Then select File > Export Selection as OGG Vorbis.

TIP: Make sure you select Export Selection as OGG Vorbis, otherwise you will be left with one file with ALL the samples playing at once, which is nasty!

2.) Next Browse to your voice folder (eg. Graeme) and then on the relevant empty ogg file listed on my example list (for example the first sample should be '2ndLeft.ogg', click save and then choose yes to overwrite the empty file.

3.) Repeat steps 1 and 2 for the rest of the sample.

This is when you will notice the benefits of recording the samples in file name order!!

TIP: Do not delete the samples in the main Audacity screen just yet, just in case something goes wrong, I would highly recommend you NEVER delete the file, instead save the whole project as an Audacity .AUP file, so you can come back to it later should you need to tinker more with any samples, for example to lengthen the gap after a sample, or re-record one because there was too much going on in the background!
4.) Once you have exported all sample, probably the most important step....CHECK YOUR FILES. Play each individual file just to make sure you've done it right! It's just too easy to make mistakes, and a mistake could be dangerous!

Step 4: Compile and Copy:

OK breathe a sigh of relief your almost there!

OK so you've exported all 59 samples into the unique ogg file names and save them in your chosen directory (For me, this was "C:\TomTom Voices\Graeme"), all you now need to do is turn this into the .chk and. vif file. And also attach .bmp (TomTom 5+ only) if you want to be snazzy!

a) Create the .chk and .vif files using win:viftool2:

win:viftool2 is a windows GUI (Graphical User Interface) alternate to the command line utility VIFTool.

Before you continue, please ensure you have installed win:viftool2

1.) Start win:viftool2 by clicking on start > programs > winviftool2 > winviftool2.

The main screen should look like this ->

3.) Click on the 'browse' button in step 1 and when the dialog box appears browse and select the folder containing your OGG files (c:\tomtom voices\graeme in my case).

When you have located the correct folder, click on ok

4.) In step two, choose a profile number (between 71 and 99 by either clicking in the number box and typing it in, or by using the plus and minus buttons.

5.) Next type in a profile name in step 3, this must be between one and eight characters long and cannot contain any illegal characters ( " / \ ? $ & ) etc.

6.) Similarly to step 1, step 4 requires you to select a folder using the 'browse' button.

Do this and navigate to a folder you wish the completed files to be stored to (for example 'desktop'), when you have done this, click ok.

7.) Once you have completed all 4 steps, click on the 'check' button. The program will now double check to make sure all the information you submitted is correct.

All going well you should see four ticks appear to the right of each step and the button 'check' should now say 'start'.

8.) Next click on start, read the warning message and then click next. The program may appear to crash momentarily then, all going well a success message will appear letting you know that the operation was completed successfully.

IMPORTANT TIPS: Your firewall program MAY show some security warnings as win:viftool2 works by compiling and running a batch file and also copies and then deletes a few temporary files. Please allow any executions to run as it will not work unless you allow it.

I can assure you this will not harm your computer!!

Please also ensure all working directories have right access, and that you are logged in with full administrative access.

When people experience problems, typically with a 'Runtime error', this is what the problem is!

b) Creating the bmp file (not essential)

You have now created the two files and are ready to create your bitmap.
The bitmap can show anything you wish, however must be exactly 160 pixels wide and 160 pixel high. It must also be called the same name as the .chk file (for example data99.bmp). You can use any type of paint programs, but the simplest would be 'Microsoft Paint'.

Once you have your three files you are now ready for the final task, copying them to your voice folder on your memory card.

c) Copying the three voice files to your memory card:

Locate to the root directory of your TomTom memory card, simply copy the three files into the following folder, depending on what version of TomTom you have:

TomTom5: Copy to the folder 'Voices' located on the 'root' of the TomTom memory card.

Pre TomTom5: Copy to the folder 'TomTom' located on the 'root' of the TomTom memory card.

IMPORTANT TIP FOR PDA TT USERS: the TomTom folder may be located on the main memory of your PDA, NOT on the memory card. If this is the case, you will need to use your synchronisation software (for example ActiveSync) and navigate to where you originally installed the main software onto.

Sada ocekujemo vase glasove - ukoliko su korektni bice postavljeni na sajtu.

Srecan rad


PS Ovo ce najverovatnije ostati najduzi single post na forumu - zato vase replike postujte u novoj temi

_________________ &
GPS navigacija, uredjaji, oprema, saveti i sve ostalo sto je vezano za GPS
Nazad na vrh
Pogledaj profil korisnika Pošalji privatnu poruku
Prikaži poruke iz poslednjih:   
Napiši novu temu   Odgovori na poruku forum forum -> GPS Software Sva vremena su GMT + 2 sata
Strana 1 od 1

Skoči na:  
Ne možete pisati nove teme u ovom forumu
Ne možete odgovarati na teme u ovom forumu
Ne možete menjati vaše poruke u ovom forumu
Ne možete brisati vaše poruke u ovom forumu
Ne možete glasati u ovom forumu

Powered by phpBB © 2001, 2005 phpBB Group
Prevod by CyberCom
Protected by Anti-Spam ACP