Program to Customize TTS Voices

Turbocc
I don't have a request for you, and I don't have a question at the moment. I just wanted to Thank you from from the members of this community for your hard work.
Your code writing is fantastic, the solutions you come up with for people are wicked fast. Please keep up the hard work, because that work does bring smiles to those of us that don't take the time to say Thanks.
And now Jill and the gang can say words Garmin never intended for them to say, or pronounce.

Thanks again

Bob

1- Which voice are you using?
2- I need you to spell Expy exactly how you see it on your screen. Make special attention to the capital letters. Ex. Expy is not the same as ExpY.

Thank you for those kind words.

I have not been posting much lately because I have a GPS overdose. I'll probably get back to it soon. I just bought another toy (digital camera) and I am spending more time on it than on my GPS. I guess I should reprogram Jill to say "Turbo! You ignore me and now you want my help?" instead of "Recalculating". LOL!

I just ran this program. It works great and it is simple to use. I was able to change "please drive the highlighted route" to "y'all follow me." Finally, my Nuvii 350 has personality.

Thank you for the cool program.

In the TTS editor Read me file it states
"So far, reported working on models 7x0, 6x0, 8x0, 3xx, 255W, 260, Zumo and
Streetpilot C550, C580 and 2730. It works too on modified StreetPilot C310, C340 and
C510."

I was wondering what the Street Pilot Mod. is?

These no modifications needed for the streetpilot unless you have a 3xx model, then TTS editor doesn't work for the reasons stated by Motorcycle Mama above.

My c550 Edited just like a nuvi with no difference,

Bob

After working my way all through this thread I was able to successfully turn Karen into a naughty navigator on my C340.
Thanks to TurboCCC and Hornbyp for their time and help.
The C340 voice file can be modified.
Chuck

This program works great. I changed it all around and got quite a few laughs from my passengers.

are you a tour coach driver?

Be aware this can void the warranty. And yes, Garmin will check software and requuire a payment before fixing it if you have installed "unauthorized" software.

Maybe Garmin will put this type of software out themselves?

Sorry to be a stick in the mud.

THIS IS WHY YOU KEEP BACK UPS OF ORG. FILES:)

You are quite correct my friend!

Also, I modified my VPM files a lot and I never crashed my Nuvi. The worst that happened was a message about corrupted audio file or no voice at all. Just copying the original VPM file back did the trick.

This is pretty harmless in any way...

Will frequent revisions using the TTS Editor on the voice files have any detremental results on the GPSr?

Also, if you revise a voice file and then run Web Updater will this affect the revised file? (This may have been covered in this thread but I can't find it.)
Chuck

I've used turbo's program from day one and it has never caused a problem on my 750. The same can't be said for the 3.2, 4.0, and 4.1 updates which Garmin released. Each of those screwed up my unit.

So I'll take any software that turbo creates over Garmin's software.

Hi Turboccc

I was using American Jill TTS. I will find out the exact abbreviation and let you know. Again, I want to thank you for such a great program.

If you modify a voice with the TTS Voice Editor, can you save the new voice as custom then transfer it to the Nuvi?

IMHO-
I would rename the modified voice file to distinguish it from the original and as not to overwrite it on the gps device.

As long as you use the same file extension, it should be recognized by the gps device.

Have you actually tried this? I'm under the impression that the file must be saved under the same name as the original. In fact, as you save the file, I thought it warned you of this.

You should use the same file name. Using a different name including changing the screen will also work as long as the original file name is renamed. Otherwise, it creates a conflict and the original voice file is used.

Long story short: use the same file name!

Using WebUpdater, the worst thing that can happen is that WebUpdater will overwrite the modified voice file in which case you can either copy the modified file back on the GPS or take the new file and modify it again.

Using TTSVoiceEditor many time on the same modified voice file does not cause problems either. I have optimized my buffer so I do not allocate more memory than necessary. In any cases, the program will let you know if you get out of buffer space. At this point, I would suggest to use the "Export" function, reload the original VPM file and do an "Import". This may help in defragmenting the buffer and increase the unused space.

I stand corrected by the creator of this great program!

thanks turboccc!

Just for fun, here are some site statistics. Below is a list of country and unique visitor (based on unique visitor per day) that went on my site. Of course, some unique visitors per day can come back the next day and will be counted again. It just gives you an idea of the number of fans per country. Impressive to see it went all over the world...

Country Unique Visitor
United States 4386
Canada 1193
United Kingdom 557
France 425
Australia 326
Germany 231
Netherlands 210
Japan 179
Belgium 125
South Africa 119
Bulgaria 115
Malaysia 86
Saudi Arabia 63
New Zealand 63
Other Countries 52
Poland 52
Italy 50
Singapore 47
Argentina 47
Norway 46
Sweden 44
Spain 43
Denmark 42
Romania 39
Thailand 39
Portugal 35
Kuwait 30
Brazil 29
Indonesia 24
Finland 23
Ireland 21
Switzerland 18
Greece 17
Croatia 17
China 16
Cyprus 12
Lithuania 9
Mexico 9
India 9
Russian Federation 9
Venezuela 8
Hungary 8
Hong Kong 8
Czech Republic 6
Slovakia 6
Latvia 4
Austria 4
Slovenia 4
Turkey 4
Tunisia 3
Bahrain 3
Malta 3
Belarus 3
Ukraine 2
Israel 2
Colombia 2
United Arab Emirates 2
Europe 1
Estonia 1
Bosnia and Herzegovina 1

First let me thank you for a very nice and useful program, it is kind of neat to be able to customize the TTS voices.

Second, I discovered by accident that the file name does not have to be 100% exactly what it was.

For example, if the original file name was "English_Australian_Karen.vpm", and the modified ones is:
"English_Australian_Karen.Modified.vpm". it still works ok, and the text between the Original name, and the ".vpm" extension can be used to identify the type of modification at the file name level.

However, there should only be one file with the original portion of the name.

Thanks again.

I used turbocc's TTS Voice Editor on my 660, and it worked great.

When I tried it on my wife's 360, the editor showed only numbers for each phrase, so I didn't know what changes I could make or what phrases the numbers referred to.

I changed the '68' line anyway. That was the same number in my 660 (numbered and identified by phrase) that referred to 'recalculating'.

This didn't have any effect on the 360.

I've recently updated the firmware on the 360.
I'm using the latest version of the editor, v2.63.
Any ideas on this?

Hi jcrayne,

Sorry to scare you like that, but NEVER CHANGE the text is there is only numbers or garbage... LOL! Hey! Hey! Seriously, no harm will come, but it will simply not work. You can always put the original voice file back and you are as good as new.

Back to the 360...

It normally works. I think I remember somebody had a similar issue. I told him to use the Garmin web updater and to update the voice file. You should try the same. You said you updated the firmware, but check again for the voice version. There may be a newer version and this one should work.

If not, just post again and I will suggest something else. Don't worry, we'll make it work. It worked for others.

Used the converter on my 760. Worked great. I just received my FREE 270 and before I dive in, I was curious if it will work on the 270? Also, does anyone know how to transfer the favorites from one unit to another?

1. The 270 doesn't have TTS voice, so the program won't work.

2. You can either copy the current.gpx file from the 760 to the 270 or you can transfer them to MapSource (or EasyGPS)from the 760 and then transfer them from there to the 270.

Thanks I will give it a try.

Got a post from jcrayne. He updated his voice files and it started to work without problems (as expected). It seems many 3x0 users have obsolete voice files. Just use the web updater and you will be good to go.

This program is great. One thing I noticed though, I modified one of the voices, and the volume seemed to be lower then the original, I went back to the program and increased the volume to about 95%, at that time when the unit spoke the modified words, the speaker sounded like it was breaking up (static). I had to go back to the original voice file!

You should have seen the look on my wife's face the first time she heard what I have Jill saying.

Can you tell me which voice? If you have the revision number of the voice file, it would help also. What gps model do you have?

A note: my program just change some ASCII characters in phrases. In no way it can affect the volume. Increasing the volume too much causes clipping in the audio and may reduce its quality.

I have a 750, and the voice I modified was English_British_Emily.vpm it is the newest version! I'll try a different voice and get back to you!

I tried the program on English Jill, I didn't have to mess with the volume, it is working fine!

Thanks for this! Just got a Nuvi 750 (my first GPS). Can't wait to start tweaking it!

I have used it before myself. Changed all kind of things. It's fun! Like when the power plug is disconnected: "Hey! Don't you think you should plug me in?" Hehe ....fun indeed!

this is also useful in the situation where you want to free some of the storage space. Back up the file, then remove any file you will not be using, such as voices, languages, graphics, etc.

WARNING:

Although the software claims on many forums to be free for non-commercial use, It's actually crippleware. (The non-TTS program quoted above, the TTS program that this thread is actually about is fine AFAIK.)

I spent hours recording and editing a new set of wav files for use with this utility. When I tried to recombine them with the utility I first got a message that I couldn't reach it's license server! What license server? It's supposedly free. So after I set up some additional security in a VM so this suspect software could phone home safely, it informed me that since I had changed over 50 wave files it would not make the vpm file without a PayPal "donation". I actually said it was going ahead and would make the file, but in a degraded mode. And sure enough, the audio seems to have a much higher than expected noise level.

Of course, any complete rebuild of the voice file is going to change over 50 wave files, The American English version of the file has 116 waves in it (the British English version actually has six less!).

So if you want to use this software to play around and edit the "recalculating" audio, fine. But if you want to make a new voice file, be warned that this is crippleware. And even if it was worth it, I would never support someone who deceptively markets his software this way.

Also I should mention a small annoyance. While you can save the vpm file with any prefix you want (such as foo.vpm), when you put it back in the Garmin it will be found under the same name as the original file it was built with. Apparently that name in in the original data header, and hidden, since I couldn't find it with a hex editor. It's sure a shame that this utility doesn't encode the new file name into the .vpm file.

You are right.

VoiceUtility which works for non-TTS voice is not 100% free. If you intend to replace all the WAV files, then it is not. There may be a limit of 50% WAV change.

My utility is TTSVoiceEditor and it works for TTS voices. It is free and will always be.

Enjoy!

TurboCCC

But I love how Australian Karen says "recalculating." It my favorite thing that she says.

I downloaded speach doctor and receive the following error when I try to run it.
"Component 'tabctl32.ocx" or one of its dependencies not currently registered: a file is missing or invalid"

I am running on windoews xp.

Does anyone have any idea what I could do to get this running?

There is a thread on this forum that addresses your problem. Search for "Component 'tabctl32.ocx" and see if it comes up. I did and found several links Turboccc is the speech doctor and creator of TTS Editor

Yup, that did the trick. I hate it when I ask a question that's already been answered.

There's so many acronyms here! I'm confused. So, with this program, can I record all my own sounds and package them as the correct file type for my Garmin 205?

Sorry. TTS refers to Text-to-Speech. The 205 does not do TTS (as far as I know). However, this program talked about here does not let you record your own voice TTS files, it just lets you change what the voices already say, but in the same voice.

This is another program that does let you record your own voice for non-TTS units (these will actually work on TTS GPSs also, but then you loose the ability to say an upcoming street name, so most TTS GPS users don't want to use them.

Unfortunately, while the program claims to be free for non-comercial use, it's a bait and switch scam, once you spend hours recording all the wave files for a new voice file and try to make a new voice file, it informs you that because you have changed over 50 sounds that it will not do a full quality conversion unless you pay, and says it's making a degraded file (and the file does sound degraded). I'm not even going to give the name of this program, as I don't want to promote someone who promotes his software this dishonest way.

So the answer to your question is that there is software to make you own file fore a nuvi 205. I'm working on if there isn't a way to get around the limitation without paying the extortion. And I sure would like to replace the American English voice with another female voice; the current one is not to my liking.

v2.64 - Added support for voice British English Serena on model 255.

Frovingslosh,

I have been hearing similar complains lately. I am 99% close to provide an alternative. I don't want to compete with his existing program because Ash10 put lots of effort in it and with the support. However, I hate to hear people missing on the good stuff because of $$$ issues. There is no real magic behind this. Just a bit of know-how...

BTW, I didn't know about the degradation. I thought it was just a number of file limitation.

For me it is not the dollar issue so much as the way that he did it. I saw the "free for non-commercial use" program and downloaded it. I spent several hours making a script of all of the American English phrases, recording them multiple times, and editing the best wave files and trimming them as close to the original trimming as I could. Ok, it turns out that I needed to trim then a bit more, but I don't hold that part against him. What I do hold against him is that I wasted hours on software that, as far as I'm concerned is bait and switch, only after spening many hours work do you find out that you have to pay up to use the software for a complete voice replacement. If I had know in advance I likely would have not wasted any time, but if the price was what I would considerable reasonable I might have paid (although never by PayPal, his only option).

Anyone has the right to build and sell software. I don't demand that it be free. I made my living in the software industry for many years. But you don't tell someone that your software is free for non-comercial use, have them spend hours working on the data for it, and then and only then tell them if they want to use it fully they have to pay up (and, adding insult to injury, they have to use PayPal).

I've sent a few freeware people donations when I thought they had done an outstanding job and I really appreciated their products. I've insisted that any freeware or shareware in commercials positions I've managed be paid for at the stated commercial rate or, if true freeware, a donation be sent or we stop using it. But I'll never give this jerk a dime. And that only show I feel about him in the mildest of terms.

The version that I used first wanted to go out to it's "license server". I wasn't happy at all about that, since there should have been no license issue for freeware and I didn't like the idea of software connecting elsewhere when not expected to. Who knows what this stranger wants to xfer out of my computer, So I wasted more time setting up a virtual machine environment so it could safely phone home without my taking a risk that someone was stealing banking, game keys or other information from my system.

After it connected to the license server, where I obviously didn't have a license, it started processing my data, but informed me that because I had 50 or more wave files I had to pay up. What nonsense is that? I can imagine someone might want to use the program to quiet an alarm beep or get rid of the annoying "recalculating", but who in their right mind would change 40 or so of the 1116 (or was it 118?) English words and not do a complete change? The software then said that it would go ahead and make the file, but do it in a degraded form until I paid the extortion.

The end result sounded pretty bad. He didn't make it clear how he was degrading things, but I think he introduced a lot of noise. Anyway, there was plenty of noise there, and I don't hear it on my original samples (although obviously I listen to my original samples and my final voice file on different hardware, so it's hard to make a comparison. The timing of my first effort has too much space between words so I think the wave files need to be trimmed better, although I'm not certain that he didn't introduce extra delays in the wave files also. I haven't bothered to find out yet, but when I get some spare time I may look into this further.

Hi Frovingslosh,

If he just introduced noise, it may be easy to get rid of it. The VPM file contains a header and a simple concatenation of all the WAV files.

Open the resulting VPM file with an HEX editor. Open also your first WAV file (something_001.WAV). Make a search for RIFF4. The first occurrence in the VPM will be be your _001.WAV. Then you can compare the, and see if it has changed. You can also compare the size of this specific WAV and see if the size was changed. If size is different, then you cannot do anything easy.

If size is the same, it is noise. You can copy the _001.WAV over the first VPM file at the proper location and you do the same for all 10x files. Ouch!

BTW, what do you have against PayPal? I use it from time to time. Any problems?

Ouch is right. Sounds like a real pain in the neck. If I get past my disgust I plan on attacking this a few other ways before I resort to that. For now I'm just disgusted with the program and have lost interest in the project.

I never used it, looked pretty hokey to me from the first day that I heard of it (back when it was a free "service"). But there are plenty of horror stories about it on the web, many people have been screwed, and I certainly don't like the way that they do business. See www.aboutpaypal.org or just www.google.com/search?q=paypal+problems for details.

I don't like the parent company either, who I consider an electronic bay of thieves who know a lot of the shilling and outright fraud is going on at their site, but don't seem to do anything at all to stop it, even when the problems are complained to them by honest users, unless the problems get so bad that they make the news. Otherwise they seem quite content to let the shill bidders go on, let the scammers pump up their seller rating with obviously false and organized "feedback", and otherwise turn a blind eye on fraud as long as they get their cut. I will never do business with them or any of their companies. I hate PayPal based on Paypal's way of treating people, but I actually might have bought Skype services if it wasn't owned by the electronic bay of thieves (not that Skype is perfect, but in comparison to the other parts of the company they are wonderful), but I can't justify providing incom to the parent company.

If people want to use it and expose themselves to the known problem, let them, but I'll have no pity on then when they get screwed. But to have that as an only option is an affront as far as I'm concerned.